| JayF | clif_h: basically under open discussion, add a link to the spec and ask, but usually folks like it to be stuck-stuck first before it hits the meeting | 00:00 |
|---|---|---|
| clif_h | sure | 00:00 |
| JayF | idk; just hard to get reviews now b/c everyone is focusing on either network isolation or the gate | 00:00 |
| *** fragatina has quit IRC | 00:04 | |
| *** rama_y has quit IRC | 00:07 | |
| *** fragatina has joined #openstack-ironic | 00:09 | |
| *** rloo has joined #openstack-ironic | 00:11 | |
| *** thrash is now known as thrash|g0ne | 00:11 | |
| *** kcalman has quit IRC | 00:13 | |
| *** fragatina has quit IRC | 00:14 | |
| *** penick has joined #openstack-ironic | 00:20 | |
| mrda | JayF: I shall use my fresh brain to full advantage | 00:29 |
| *** joprovost1 has joined #openstack-ironic | 00:40 | |
| *** mtanino has quit IRC | 00:41 | |
| *** joprovost has quit IRC | 00:42 | |
| *** joprovost1 is now known as joprovost | 00:42 | |
| *** rloo has quit IRC | 00:45 | |
| *** baoli has joined #openstack-ironic | 00:45 | |
| *** lifeless has quit IRC | 00:46 | |
| *** lifeless has joined #openstack-ironic | 00:47 | |
| *** derekh has joined #openstack-ironic | 00:47 | |
| *** rcernin has quit IRC | 00:54 | |
| *** joprovost has quit IRC | 00:55 | |
| *** aNupoisc has quit IRC | 00:55 | |
| *** [1]cdearborn has joined #openstack-ironic | 00:57 | |
| *** mbound has quit IRC | 00:59 | |
| *** absubram has joined #openstack-ironic | 01:01 | |
| *** absubram_ has joined #openstack-ironic | 01:02 | |
| *** absubram has quit IRC | 01:05 | |
| *** absubram_ is now known as absubram | 01:05 | |
| *** derekh has quit IRC | 01:09 | |
| *** ijw has joined #openstack-ironic | 01:16 | |
| *** rloo has joined #openstack-ironic | 01:19 | |
| *** rama_y has joined #openstack-ironic | 01:22 | |
| *** shakamunyi has joined #openstack-ironic | 01:23 | |
| *** ijw has quit IRC | 01:23 | |
| *** penick has quit IRC | 01:27 | |
| *** rbudden has joined #openstack-ironic | 01:28 | |
| *** jaybeale has joined #openstack-ironic | 01:30 | |
| *** jaybeale has quit IRC | 01:38 | |
| *** jaybeale has joined #openstack-ironic | 01:39 | |
| *** jaybeale has quit IRC | 01:40 | |
| *** jaybeale has joined #openstack-ironic | 01:40 | |
| *** Goneri has quit IRC | 01:41 | |
| *** jaybeale has quit IRC | 01:42 | |
| *** rloo has quit IRC | 01:42 | |
| *** rloo has joined #openstack-ironic | 01:43 | |
| *** rloo has quit IRC | 01:43 | |
| *** amotoki has joined #openstack-ironic | 01:46 | |
| *** joprovost has joined #openstack-ironic | 01:51 | |
| *** yuanying has quit IRC | 01:51 | |
| *** yuanying has joined #openstack-ironic | 01:52 | |
| *** yuanying has quit IRC | 01:52 | |
| *** yuanying has joined #openstack-ironic | 01:53 | |
| *** Goneri has joined #openstack-ironic | 01:54 | |
| *** [1]cdearborn has quit IRC | 01:57 | |
| *** mbound has joined #openstack-ironic | 02:00 | |
| *** mbound has quit IRC | 02:04 | |
| *** hoangcx has joined #openstack-ironic | 02:08 | |
| *** Goneri has quit IRC | 02:20 | |
| tonyb | who here knows about pyghmi? | 02:21 |
| mrda | tonyb: I think you're asking at the wrong time of day | 02:22 |
| tonyb | mrda: It's always the wrong TOD :( | 02:23 |
| mrda | tonyb: welcome to my world :) | 02:23 |
| *** garthb has joined #openstack-ironic | 02:25 | |
| *** penick has joined #openstack-ironic | 02:37 | |
| joprovost | lâche pas | 02:38 |
| *** daemontool has quit IRC | 02:38 | |
| joprovost | oops, wrong window | 02:38 |
| *** garthb has quit IRC | 02:39 | |
| *** baoli has quit IRC | 02:51 | |
| *** rbudden has quit IRC | 02:51 | |
| *** yuanying has quit IRC | 03:00 | |
| *** Sukhdev has quit IRC | 03:02 | |
| *** yuanying has joined #openstack-ironic | 03:24 | |
| *** sacharya has quit IRC | 03:31 | |
| *** links has joined #openstack-ironic | 03:34 | |
| *** irf has joined #openstack-ironic | 03:41 | |
| *** yuanying has quit IRC | 03:42 | |
| *** fragatina has joined #openstack-ironic | 03:43 | |
| *** yuanying has joined #openstack-ironic | 03:44 | |
| *** fragatina has quit IRC | 03:45 | |
| *** fragatina has joined #openstack-ironic | 03:46 | |
| *** fragatina has quit IRC | 03:46 | |
| *** fragatina has joined #openstack-ironic | 03:47 | |
| *** yuanying has quit IRC | 04:01 | |
| *** Fdaisuke_ has joined #openstack-ironic | 04:02 | |
| *** Fdaisuke has quit IRC | 04:04 | |
| *** yuanying has joined #openstack-ironic | 04:07 | |
| *** vmud213 has joined #openstack-ironic | 04:22 | |
| openstackgerrit | Sivaramakrishna Garimella proposed openstack/ironic: Add portgroups to support LAG interfaces - API https://review.openstack.org/206244 | 04:24 |
| *** pcaruana has joined #openstack-ironic | 04:25 | |
| *** yuanying has quit IRC | 04:27 | |
| *** harlowja_at_home has joined #openstack-ironic | 04:28 | |
| *** sacharya has joined #openstack-ironic | 04:32 | |
| *** pcaruana has quit IRC | 04:32 | |
| *** yuanying has joined #openstack-ironic | 04:34 | |
| *** yuanying has quit IRC | 04:36 | |
| *** yuanying has joined #openstack-ironic | 04:37 | |
| *** sacharya has quit IRC | 04:37 | |
| *** yuanying has quit IRC | 04:39 | |
| *** yuanying has joined #openstack-ironic | 04:41 | |
| *** yuanying has quit IRC | 04:42 | |
| *** yuanying has joined #openstack-ironic | 04:43 | |
| *** appu103_ has joined #openstack-ironic | 04:45 | |
| *** joprovost has quit IRC | 04:46 | |
| *** appu103_ is now known as aparnav | 04:47 | |
| *** yuanying has quit IRC | 04:48 | |
| *** yuanying has joined #openstack-ironic | 04:51 | |
| *** yuanying has quit IRC | 04:53 | |
| *** yuanying has joined #openstack-ironic | 04:54 | |
| *** Fdaisuke_ has quit IRC | 04:55 | |
| *** Sukhdev has joined #openstack-ironic | 04:56 | |
| *** Fdaisuke_ has joined #openstack-ironic | 04:56 | |
| *** yuanying has quit IRC | 05:00 | |
| *** penick has quit IRC | 05:00 | |
| *** penick has joined #openstack-ironic | 05:03 | |
| *** rama_y has quit IRC | 05:16 | |
| *** rama_y has joined #openstack-ironic | 05:17 | |
| *** yuanying has joined #openstack-ironic | 05:18 | |
| *** stendulker has joined #openstack-ironic | 05:19 | |
| *** yuanying has quit IRC | 05:23 | |
| *** yuanying has joined #openstack-ironic | 05:25 | |
| openstackgerrit | Dan Bode proposed openstack/bifrost: Document that ssh_public_key_path must be set https://review.openstack.org/312782 | 05:27 |
| *** yuanying has quit IRC | 05:31 | |
| *** harlowja_at_home has quit IRC | 05:31 | |
| *** yuanying has joined #openstack-ironic | 05:32 | |
| *** yuanying has quit IRC | 05:33 | |
| *** yuanying has joined #openstack-ironic | 05:35 | |
| *** amotoki_ has joined #openstack-ironic | 05:39 | |
| *** ChubYann has quit IRC | 05:40 | |
| *** rama_y has quit IRC | 05:43 | |
| *** amotoki has quit IRC | 05:43 | |
| *** rcernin has joined #openstack-ironic | 05:54 | |
| *** yuanying has quit IRC | 05:55 | |
| *** rama_y has joined #openstack-ironic | 05:56 | |
| *** yuanying has joined #openstack-ironic | 05:57 | |
| *** yuanying has quit IRC | 06:03 | |
| *** yuanying has joined #openstack-ironic | 06:04 | |
| *** pcaruana has joined #openstack-ironic | 06:12 | |
| *** rama_y has quit IRC | 06:19 | |
| *** jcoufal has joined #openstack-ironic | 06:26 | |
| *** sacharya has joined #openstack-ironic | 06:33 | |
| *** YiboCai has joined #openstack-ironic | 06:34 | |
| *** sacharya has quit IRC | 06:38 | |
| *** hoangcx has quit IRC | 06:50 | |
| *** thiagop has quit IRC | 06:50 | |
| *** xavierr has quit IRC | 06:52 | |
| *** xavierr has joined #openstack-ironic | 06:53 | |
| *** thiagop has joined #openstack-ironic | 06:53 | |
| *** crinkle_ has joined #openstack-ironic | 06:59 | |
| *** Sukhdev has quit IRC | 06:59 | |
| *** crinkle has quit IRC | 07:00 | |
| *** crinkle_ is now known as crinkle | 07:00 | |
| *** mmedvede has quit IRC | 07:00 | |
| *** elfosardo has quit IRC | 07:00 | |
| *** lennyb has quit IRC | 07:03 | |
| *** tesseract has joined #openstack-ironic | 07:05 | |
| *** tesseract has quit IRC | 07:09 | |
| *** elfosardo has joined #openstack-ironic | 07:13 | |
| milan_ | morning Ironic! :) | 07:19 |
| *** daemontool has joined #openstack-ironic | 07:19 | |
| *** ChrisAusten has quit IRC | 07:20 | |
| *** e0ne has joined #openstack-ironic | 07:23 | |
| *** chopmann has joined #openstack-ironic | 07:24 | |
| *** e0ne has quit IRC | 07:26 | |
| *** mmedvede has joined #openstack-ironic | 07:27 | |
| *** sacharya has joined #openstack-ironic | 07:34 | |
| *** sacharya has quit IRC | 07:39 | |
| *** dmk0202 has joined #openstack-ironic | 07:40 | |
| pas-ha | morning all | 07:51 |
| *** fragatina has quit IRC | 07:56 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-ironic | 08:01 | |
| *** ifarkas has joined #openstack-ironic | 08:08 | |
| *** kromanenko_ has joined #openstack-ironic | 08:12 | |
| vsaienko | morning pas-ha, Ironic'ers | 08:14 |
| *** athomas has joined #openstack-ironic | 08:15 | |
| *** hoangcx has joined #openstack-ironic | 08:16 | |
| *** dtantsur|afk is now known as dtantsur | 08:17 | |
| dtantsur | Morning Ironic | 08:17 |
| openstackgerrit | Nisha Agarwal proposed openstack/proliantutils: [WIP] Test patch for gpu device count https://review.openstack.org/315402 | 08:17 |
| *** ohamada has joined #openstack-ironic | 08:17 | |
| milan_ | dtantsur, vsaienko pas-ha morning guys! | 08:17 |
| *** lucas-dinner is now known as lucasagomes | 08:17 | |
| lucasagomes | morning all | 08:18 |
| milan_ | lucasagomes, o/ :) | 08:18 |
| betherly | morning! | 08:20 |
| milan_ | betherly, o/ | 08:25 |
| irf | morning Ironic | 08:25 |
| *** heyongli has quit IRC | 08:30 | |
| *** heyongli has joined #openstack-ironic | 08:31 | |
| *** penick has quit IRC | 08:33 | |
| *** penick has joined #openstack-ironic | 08:34 | |
| openstackgerrit | Grzegorz Grasza (xek) proposed openstack/ironic-inspector-specs: Generic PCI Resource spec https://review.openstack.org/315413 | 08:34 |
| *** jistr has joined #openstack-ironic | 08:35 | |
| dtantsur | xek, hi! how does https://bugs.launchpad.net/ironic-inspector/+bug/1580893 play with https://bugs.launchpad.net/ironic-python-agent/+bug/1571580 ? | 08:37 |
| openstack | Launchpad bug 1580893 in Ironic Inspector "[RFE] Generic PCI Resource" [Undecided,In progress] - Assigned to Grzegorz Grasza (xek) | 08:37 |
| openstack | Launchpad bug 1571580 in ironic-python-agent "[RFE] In-band capabilities discovery with IPA and inspector" [Wishlist,In progress] - Assigned to Dmitry Tantsur (divius) | 08:37 |
| * dtantsur just saw the spec patch posted | 08:37 | |
| *** ohamada has quit IRC | 08:38 | |
| *** ohamada has joined #openstack-ironic | 08:38 | |
| irf | issue got resolved .... | 08:38 |
| irf | very happy now :) | 08:39 |
| dtantsur | congrats :) | 08:39 |
| *** e0ne has joined #openstack-ironic | 08:39 | |
| xek | dtantsur, Hi I talked with you about this after your presentation at the summit | 08:39 |
| dtantsur | yeah. a bit hard to remember right now :) | 08:40 |
| xek | dtantsur, this is specifically to enable generic detection of PCI devices | 08:40 |
| irf | actually, only that error has resolved now ... still need to make changes to provision the machine and get going .. long way to go still :) | 08:40 |
| dtantsur | xek, aha, so it's supposed to be built on https://bugs.launchpad.net/ironic-python-agent/+bug/1571580 right? | 08:40 |
| openstack | Launchpad bug 1571580 in ironic-python-agent "[RFE] In-band capabilities discovery with IPA and inspector" [Wishlist,In progress] - Assigned to Dmitry Tantsur (divius) | 08:40 |
| xek | dtantsur, yes, maybe (I'm reading your spec) | 08:41 |
| *** oberezovskyi has joined #openstack-ironic | 08:41 | |
| oberezovskyi | Hi! I see, that there is no ./lib/ironic in devstack on mitaka. Devstack+Ironic installation is not available yet? | 08:42 |
| openstackgerrit | Merged openstack/virtualbmc: Add unittests for the manager.py module https://review.openstack.org/314681 | 08:44 |
| vsaienko | oberezovskyi: Ironic uses devstack plugin in Mitaka, the code has been moved to ironic repository https://github.com/openstack/ironic/tree/master/devstack | 08:45 |
| oberezovskyi | vsaienko: thx for information! | 08:46 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: TEST Checking Gates https://review.openstack.org/311109 | 08:50 |
| *** electrofelix has joined #openstack-ironic | 08:54 | |
| xek | dtantsur, ok, I think I it's a good idea to implement pci device detection as part of the capabilities plugin, is it also what you think? | 08:54 |
| dtantsur | xek, exactly | 08:54 |
| *** pal_bth has joined #openstack-ironic | 08:55 | |
| dtantsur | I want to propose code for inspector part today btw | 08:56 |
| dtantsur | IPA part is already up: https://review.openstack.org/308292 | 08:57 |
| xek | dtantsur, cool, I'l have a look :) | 08:57 |
| pal_bth | Hello, I've tried to figure out if Ironic will allow you to 'save' a deployed node? I.e. allow you to do a copy/backup of it, stored within OS framework, for deployment again later. | 09:00 |
| *** pcaruana is now known as pcaruana|afk| | 09:01 | |
| *** oberezovskyi has left #openstack-ironic | 09:02 | |
| milan_ | dtantsur, lucasagomes: the ipxe chainloading bz: https://bugzilla.redhat.com/show_bug.cgi?id=1335440 | 09:03 |
| openstack | bugzilla.redhat.com bug 1335440 in ipxe "chainloading big ipxe config file causes boot to hang while booting through http" [Unspecified,New] - Assigned to virt-maint | 09:03 |
| lucasagomes | milan_, thanks! | 09:03 |
| * lucasagomes looks | 09:03 | |
| dtantsur | thanks | 09:03 |
| milan_ | np | 09:03 |
| dtantsur | pal_bth, not right now. I think we have a spec up for that. | 09:03 |
| *** jtomasek_ has joined #openstack-ironic | 09:04 | |
| *** openstackgerrit has quit IRC | 09:04 | |
| *** openstackgerrit has joined #openstack-ironic | 09:05 | |
| pal_bth | Ok, dtantsur thanks, is it possible to see that spec? | 09:05 |
| pal_bth | What's it called? | 09:06 |
| lucasagomes | pal_bth, is it a cold snapshot? (e.g, stop the node, boot something that will copy the image to glance) | 09:07 |
| pal_bth | Yes lucasagomes. | 09:08 |
| lucasagomes | pal_bth, right, not sure we do have a spec, will look, but it's more complicated than it appears to be when you have things like configdrvice and/or network isolation | 09:10 |
| lucasagomes | configdrive because the way Ironic does is by having a partition on the disk, and that partition will have the secrets specific for that instance | 09:11 |
| milan_ | lucasagomes, dtantsur wdyt? is the bug lgtm? | 09:11 |
| lucasagomes | network isolation because the tenants network won't have access to the control plane | 09:11 |
| lucasagomes | milan_, it's interesting, funny tho, the mac specific files are longer and are also script files being fetched via http | 09:12 |
| milan_ | lucasagomes, could be something within the chain-load ipxe command | 09:13 |
| *** oberezovskyi has joined #openstack-ironic | 09:14 | |
| lucasagomes | vdrok, pas-ha re resource tracker, I have a devstack setup here (fresh, installed yesterday) but I can't reproduce that behavior http://paste.openstack.org/show/496859/ | 09:14 |
| lucasagomes | milan_, yeah, very interesting bug | 09:14 |
| lucasagomes | milan_, also... keep in mind that testing on VMs is not using the iPXE ROM in the tftp | 09:15 |
| lucasagomes | it's using the ones embedded in qemu, which can be very old | 09:15 |
| pas-ha | lucasagomes: the pastebin is for what you see? because it perfectly reporduces - memory_mb and local_gb are zero | 09:16 |
| lucasagomes | pas-ha, oh right yes | 09:17 |
| dtantsur | sambetts, reported https://bugs.launchpad.net/ironic-python-agent/+bug/1580932 so that we don't forget | 09:17 |
| openstack | Launchpad bug 1580932 in ironic-python-agent "The 'logs' inspection collector does not work with TinyIPA" [Low,Triaged] | 09:17 |
| lucasagomes | pas-ha, sorry I was looking at the other fields | 09:17 |
| * lucasagomes needs more coffee apparently | 09:17 | |
| lucasagomes | pas-ha, thanks I will investigate that | 09:17 |
| pas-ha | :) | 09:17 |
| pas-ha | we found an offending patch in Nova btw, is mentioned now in the bug | 09:18 |
| *** sivaramakrishna has joined #openstack-ironic | 09:19 | |
| sivaramakrishna | Morning ironicers, | 09:19 |
| lucasagomes | nice | 09:20 |
| sivaramakrishna | Hi lucasogomes, | 09:20 |
| pal_bth | lucasagomes i did not expect it to be easy : ) but I think it would be possible, either in simple cases two images same hardware, or slightly more complicated cases would require/gain to interact with the networking components that could wrap the network environment for the node.. | 09:21 |
| lucasagomes | sivaramakrishna, good morning | 09:21 |
| sivaramakrishna | hi.. get this error http://paste.openstack.org/show/496860/ | 09:21 |
| lucasagomes | pal_bth, oh totally possible | 09:21 |
| pal_bth | miliage just varies :) | 09:21 |
| sivaramakrishna | jenkins is failing in tempest driver tests.. | 09:21 |
| sivaramakrishna | for https://review.openstack.org/#/c/206244/ | 09:22 |
| sivaramakrishna | any idea? | 09:22 |
| *** jtomasek has quit IRC | 09:22 | |
| lucasagomes | pal_bth, yes, and in the future it would be very interesting to play with btrfs snapshots | 09:23 |
| pal_bth | lucasagomes, in the future I'd like to move a node to and from hypervisors and BM as needed ... But the first part, the saving, no specification yet. Possible plans? | 09:25 |
| vdrok | morning ironic, lucasagomes dtantsur milan_ ! | 09:25 |
| lucasagomes | pal_bth, yeah no specs that I can find | 09:26 |
| sivaramakrishna | good morning vdrok! | 09:26 |
| lucasagomes | pal_bth, do you feel like writing one for it? | 09:26 |
| vdrok | morning sivaramakrishna | 09:26 |
| sambetts | Morning all | 09:27 |
| pal_bth | Possibly, how and where? :) | 09:27 |
| sambetts | dtantsur: awesome thanks for creating that bug repotr | 09:27 |
| lucasagomes | pal_bth, http://docs.openstack.org/developer/ironic/dev/code-contribution-guide.html?highlight=rfe#adding-new-features | 09:27 |
| sivaramakrishna | getting this error http://paste.openstack.org/show/496860/ for #206244 | 09:27 |
| sivaramakrishna | any idea? | 09:28 |
| pal_bth | lucasagomes Thanks, will look into it, and get back if I have questions. | 09:28 |
| lucasagomes | pal_bth, +1, thanks! | 09:28 |
| irf | ok now .. moving further, ran the command : "nova boot --flavor baremetal --image cirros --min-count 1 --nic net-id=3ea58577-f3cf-418b-8915-6cc7fc13c628 test", node is getting booted and getting IP also from DHCP.. and now stuck on TFTP open timeout error .... | 09:30 |
| pal_bth | lucasagomes; this would (initially) be a RFE right? | 09:32 |
| lucasagomes | pal_bth, yeah, but I believe that will require a spec | 09:32 |
| openstackgerrit | Sharat Sharma proposed openstack/ironic: Added the request parameter maintenance for node.update. https://review.openstack.org/314917 | 09:32 |
| lucasagomes | pal_bth, for changes that are really straight forward a spec is not required, but this have some corner cases such as the configdrive and network isolation that we talked about briefly | 09:33 |
| lucasagomes | that may require a spec, IMO | 09:33 |
| vdrok | morning sambetts | 09:33 |
| sambetts | o/ vdrok | 09:33 |
| *** sacharya has joined #openstack-ironic | 09:35 | |
| *** jtomasek_ is now known as jtomasek | 09:35 | |
| pal_bth | Definitely, lucasagomes, I'll look at the approved specs., etc, and get back. | 09:36 |
| lucasagomes | cool | 09:36 |
| *** sacharya has quit IRC | 09:39 | |
| milan_ | morning vdrok sambetts :) | 09:42 |
| sambetts | o/ milan_, you've gained a _ | 09:43 |
| milan_ | yeah, my irc client was so generous as to give me one ;) | 09:46 |
| milan_ | but I bet dtantsur will blame my OS ;) | 09:46 |
| dtantsur | of course ;) | 09:46 |
| sambetts | heh :P | 09:46 |
| sambetts | milan_: /msg NickServ REGAIN milan | 09:46 |
| pal_bth | lucasgomes; is this a acceptable starting view of the current system? http://docs.openstack.org/developer/ironic/deploy/user-guide.html | 09:46 |
| pal_bth | i,e, either matches current reality, or close enough? | 09:47 |
| milan_ | sambetts, no work | 09:47 |
| * milan_ relocates see ya all in a bit | 09:48 | |
| sambetts | milan_: :/ | 09:48 |
| *** milan_ has quit IRC | 09:49 | |
| *** hshiina has joined #openstack-ironic | 09:49 | |
| *** amotoki has joined #openstack-ironic | 09:52 | |
| lucasagomes | pal_bth, yes that does, but it's very high level | 09:53 |
| * lucasagomes missed the ping because there's an "a" between lucas and gomes, I was thinking about changing my IRC handle actually... it's too long | 09:54 | |
| *** amotoki_ has quit IRC | 09:54 | |
| * lucasagomes has "lcs" registered at freenode | 09:54 | |
| pal_bth | lucasagomes Any mid-level documentation more suitable? (sorry for missing the ping, had the url in the pastebin). | 09:54 |
| sambetts | lucasagomes: could just make it lucas | 09:54 |
| *** mgould has joined #openstack-ironic | 09:55 | |
| lucasagomes | pal_bth, if that helps, there are a some youtube videos form devananda explaining the ironic structure | 09:55 |
| * lucasagomes hates searching for ironic and matching a lot of Alanis Morissette | 09:55 | |
| sambetts | XD | 09:55 |
| mgould | morning everyone! | 09:56 |
| lucasagomes | pal_bth, this is a good one https://www.youtube.com/watch?v=vjGPuBfZ87A | 09:56 |
| lucasagomes | sambetts, I think lucas is already registered | 09:56 |
| lucasagomes | would be a surprise if it's not | 09:56 |
| * lucasagomes checks | 09:56 | |
| mgould | lucasagomes, I should really watch that video some time... | 09:56 |
| lucasagomes | mgould, ++ | 09:57 |
| pal_bth | lucasagomes would be great, searched youtube for devananda, did not match what I expected. thanks for the url | 09:57 |
| lucasagomes | yw | 09:57 |
| sambetts | mgould: morning | 09:58 |
| sambetts | lucasagomes: yeah I guess its quite likely its taken :/ | 09:58 |
| lucasagomes | sambetts, yup [lucas] (~lucas@xanadu.blop.info): Lucas Nussbaum | 09:59 |
| * lucasagomes should go for lcs anyway | 10:00 | |
| sambetts | :-P yeah thats pretty neat | 10:00 |
| sivaramakrishna | hi vdrok, | 10:01 |
| sivaramakrishna | sambetts | 10:02 |
| sambetts | o/ sivaramakrishna | 10:02 |
| sivaramakrishna | jenkins failing for some reason on https://review.openstack.org/#/c/206244/ | 10:03 |
| sivaramakrishna | I'm not exactly sure why.. | 10:03 |
| mgould | morning sambetts | 10:03 |
| sivaramakrishna | any idea? | 10:03 |
| mgould | also lucasagomes pal_bth sivaramakrishna vdrok dtantsur and anyone else I've missed... | 10:03 |
| vdrok | morning mgould :) | 10:03 |
| sivaramakrishna | I pasted an error in http://paste.openstack.org/show/496860/ of the nova compute error | 10:04 |
| sivaramakrishna | is it something to do with DB? | 10:04 |
| pal_bth | mgould good morning, lunch or evening depending on where you are, anyway have a nice one. | 10:05 |
| sivaramakrishna | sambetts,lucasagomes,vdrok: seems main issue is that compute record isn't getting created.. | 10:05 |
| sambetts | sivaramakrishna: that looks like something has gone wrong with your DB :/ | 10:06 |
| sivaramakrishna | I haven't modified the db code though.. | 10:07 |
| mgould | pal_bth, late morning here, but we observe http://www.total-knowledge.com/~ilya/mips/ugt.html :-) | 10:07 |
| pal_bth | mgould thanks for the pointer. | 10:08 |
| sivaramakrishna | sambetts: does it have anything to do with https://review.openstack.org/#/c/259470/? | 10:08 |
| sambetts | that was merged 3 months ago removing support for compute nodes created on a kilo OpenStack | 10:09 |
| sivaramakrishna | hm.. true | 10:09 |
| sivaramakrishna | still not able to get to the issue, as patch 96 passed jenkins | 10:10 |
| sivaramakrishna | oops patch 94 rather | 10:11 |
| sivaramakrishna | sambetts:the differences I see is in https://review.openstack.org/#/c/206244/94..101/ironic/conductor/manager.py | 10:15 |
| sivaramakrishna | mainly related to clean states | 10:15 |
| sivaramakrishna | cleaning steates | 10:15 |
| *** YiboCai has quit IRC | 10:16 | |
| irf | hello, what cud be the reason for TFTP open timeout ?? i followed the instructions : http://docs.openstack.org/developer/ironic/deploy/install-guide.html#pxe-setup | 10:17 |
| irf | is it the case that , node is not able to access tftproot folder ? | 10:17 |
| sambetts | it looks like a difference in nova thats causing issues, I don't think that the Ironic changes are causing the problem, I'd check which version of nova is installed in each of those runs | 10:18 |
| sivaramakrishna | how to check that? | 10:18 |
| sivaramakrishna | sambetts: will be back in a while.. | 10:18 |
| sambetts | sivaramakrishna: there are quite often pip freeze outputs in test logs you can see which version is in there | 10:19 |
| dtantsur | sambetts, any ideas about https://bugs.launchpad.net/ironic-python-agent/+bug/1559448 ? | 10:19 |
| openstack | Launchpad bug 1559448 in ironic-python-agent "tinyipa build failed" [Undecided,New] - Assigned to Haomeng,Wang (whaom) | 10:19 |
| sambetts | or if you have access to the machine, cd /opt/stack/noval git log | 10:19 |
| irf | yes the nova version is different on both the nodes .. controller and compute | 10:19 |
| irf | on controller it is : 3.2.0 and on compute 2.30.1 | 10:20 |
| sambetts | irf: that was @ sivaramakrishna | 10:20 |
| irf | ohhh | 10:21 |
| irf | :) | 10:21 |
| irf | so any suggesions for that error ? | 10:21 |
| sambetts | irf: does your node thats failing tftp have a valid network route to the tftp server? | 10:21 |
| irf | i assume tftp server is the controller node and yes, both the ironic node and controller are on same network | 10:22 |
| irf | i have configured the tftp on the controller node | 10:23 |
| sambetts | is the TFTP IP address being passed to the node correct? | 10:23 |
| *** hoangcx has quit IRC | 10:23 | |
| irf | i guess yes.. i can boot the ironic node using command : nova boot --flavor baremetal --image cirros --min-count 1 --nic net-id=3ea58577-f3cf-418b-8915-6cc7fc13c628 test | 10:24 |
| sambetts | dtantsur: I think there are 2 issues here, the first is that ideally you shouldn't run it as root, and I think the first log there seems to be that sudo is behaving weird on this guys machine, the second log running sudo make instead, is that this guys locale isn't supported when pip installing pbr, it maybe that we need to enforce a locale or he needs to switch his locale or something :/ | 10:26 |
| dtantsur | sambetts, could you please follow-up on a bug? | 10:27 |
| sambetts | dtantsur: yeah sure :) | 10:27 |
| * dtantsur does a spring bug cleaning :) | 10:27 | |
| dtantsur | jroll, hi! should I add you as an assignee for https://bugs.launchpad.net/ironic/+bug/1526401 ? | 10:29 |
| openstack | Launchpad bug 1526401 in Ironic "[RFE] Pluggable network providers" [Wishlist,In progress] | 10:29 |
| *** pal_bth has quit IRC | 10:31 | |
| irf | is there any comamnd to check wheter tftp server is running ? | 10:32 |
| mgould | irf, `pgrep dnsmasq` | 10:33 |
| irf | i ran the comamnd : ps -ef| grep tftp | 10:33 |
| irf | output : /usr/sbin/in.tftpd --listen --user tftp --address [::]:69 --secure /var/lib/tftpboot | 10:33 |
| mgould | yeah, dnsmasq does its own TFTP serving | 10:33 |
| *** pal_bth has joined #openstack-ironic | 10:33 | |
| mgould | oh, huh | 10:33 |
| irf | ok so it means tftp servr is runnig .... | 10:34 |
| mgould | well, it means *a* tftp server is running | 10:34 |
| mgould | though it only appears to be listening on IPv6 | 10:34 |
| mgould | can you run `pgrep -a dnsmasq`? | 10:35 |
| mgould | also, I'd expect it to be looking for images in /tftpboot | 10:35 |
| mgould | but perhaps that's only true of ironic-inspector's dnsmasq | 10:35 |
| irf | http://pastebin.com/uwZZrTX8 | 10:39 |
| mgould | huh, that's a lot of dnsmasq processes | 10:41 |
| vdrok | irf: have you restarted xinetd after changing all that is needed? If so, you should have in.tftpd using /tftpboot, not /var/lib/tftpboot | 10:41 |
| *** alexpilotti has joined #openstack-ironic | 10:42 | |
| mgould | vdrok, does neutron use xinetd for TFTP service? | 10:42 |
| *** MattMan has quit IRC | 10:42 | |
| *** MattMan has joined #openstack-ironic | 10:42 | |
| vdrok | mgould: I think we do, not neutron :) http://docs.openstack.org/developer/ironic/deploy/install-guide.html#pxe-setup | 10:43 |
| mgould | vdrok, thanks | 10:43 |
| vdrok | neutron uses dnsmasq, but I don't know if it does anything with tftp | 10:44 |
| mgould | thanks | 10:44 |
| mgould | IIRC we use dnsmasq as a tftp server in ironic-inspector | 10:44 |
| vdrok | yep , I did that for standalone setup too | 10:45 |
| dtantsur | neutron itself does not manage tftp, we do. | 10:45 |
| dtantsur | we only both DHCP flags to neutron | 10:45 |
| sivaramakrishna | sambetts:you mean to say.. jenkins failing because of some recent merge of nova? | 10:45 |
| sambetts | sivaramakrishna: not sure without digging deeper, but thats where the big error is you pasted was | 10:47 |
| sivaramakrishna | oh ok.. I pasted it from nova compute log of that run | 10:47 |
| sivaramakrishna | should I do a recheck then? | 10:48 |
| vdrok | sivaramakrishna: nope, please wait for now | 10:48 |
| vdrok | sambetts: I see this on other recent runs too http://logs.openstack.org/17/314917/2/check/gate-tempest-dsvm-ironic-pxe_ssh/71d0687/logs/screen-n-cpu.txt.gz#_2016-05-12_10_09_27_393 | 10:48 |
| vdrok | so it should be something merged yesterday/today | 10:49 |
| sambetts | vdrok: yeah looks like nova derped something maybe | 10:49 |
| sambetts | vdrok: same thing is causing my CI to fail | 10:50 |
| irf | yes i have restared the xinted service .. let me chk again | 10:53 |
| irf | in the file : /etc/xinetd.d/tftp, i have given the path as /tftpboot and not the /var/lib/tftpboot | 10:59 |
| sivaramakrishna | it looks like the record is created initially..2016-05-12 04:51:21.992 14560 INFO nova.compute.resource_tracker [req-879d8b78-a351-4771-a6de-b459cfbeae63 - -] Compute_service record updated for ubuntu-trusty-rax-ord-708340:afba5427-7bdf-42a0-b357-1e9472582205 | 11:00 |
| vdrok | dtantsur: lucasagomes sambetts looking at logstash, this is it - https://review.openstack.org/#/c/312750/ | 11:00 |
| vdrok | causing OrphanedObjectError: Cannot call obj_load_attr on orphaned Instance object | 11:01 |
| vdrok | message: "OrphanedObjectError" AND module: "nova.virt.ironic.driver" | 11:01 |
| dtantsur | meh.... | 11:02 |
| sambetts | -,- | 11:02 |
| * lucasagomes looks | 11:02 | |
| lucasagomes | *sigh* | 11:03 |
| irf | ok.. i copied all contents of /tftpboot to /var/lib/tftpboot | 11:03 |
| irf | and cganged the owner of the dir to ironic | 11:03 |
| irf | changed* | 11:03 |
| irf | restarted the xinetd service | 11:03 |
| *** sivaramakrishna has quit IRC | 11:04 | |
| *** amotoki has quit IRC | 11:06 | |
| *** vmud213 has quit IRC | 11:08 | |
| *** amotoki has joined #openstack-ironic | 11:10 | |
| *** mgould has quit IRC | 11:12 | |
| *** hamzy has quit IRC | 11:12 | |
| jroll | tonyb: depends what you want to know about pyghmi :) jjohn<tab> is the expert, a few of us know some things about it. | 11:12 |
| jroll | dtantsur: sure | 11:12 |
| irf | tftp server shud be run by ironic user ? | 11:12 |
| jroll | morning everyone :) | 11:12 |
| tonyb | jroll: I just wanted to know it 1.0.0 was a API break and should we think about capping than on stable/* | 11:13 |
| vdrok | morning jroll, we might have another issue with nova, see ^^ or on whiteboard | 11:13 |
| jroll | vdrok: FUN | 11:13 |
| tonyb | jroll: ISTR that virtualbmc didn't work with 1.0.0 when it came out ... | 11:13 |
| jroll | vdrok: I think that's just a matter of missing a context | 11:14 |
| vdrok | yeah, seems so | 11:14 |
| jroll | tonyb: ah, I don't think it was an intentional break, lucasagomes do you have any thoughts? | 11:14 |
| *** amotoki has quit IRC | 11:14 | |
| *** dansmith has quit IRC | 11:14 | |
| jroll | tonyb: fwiw, we don't put pyghmi in reqs.txt, fwiw, it's optional | 11:14 |
| irf | now , it is runnng by root user | 11:15 |
| tonyb | jroll: Yeah I saw that ;P | 11:15 |
| tonyb | jroll: it caused me a little confusion whiel I was trying to find "thing we dont use" in g-r ;P | 11:15 |
| jroll | tonyb: yeah :/ | 11:16 |
| dtantsur | morning jroll | 11:16 |
| jroll | ohai | 11:16 |
| sambetts | o/ jroll | 11:17 |
| *** dansmith has joined #openstack-ironic | 11:17 | |
| dtantsur | jroll, also: is it possible to release IPA from Mitaka somewhere in the near future? | 11:17 |
| *** dansmith is now known as Guest51926 | 11:17 | |
| jroll | dtantsur: totally, feel free to propose one whenever, I can ack | 11:17 |
| dtantsur | good | 11:17 |
| dtantsur | I will probably do it today then. I'd like these bulky inspection fixes to get released. | 11:18 |
| jroll | vdrok: patch incoming shortly | 11:20 |
| vdrok | jroll: oh, great, thanks! :) | 11:21 |
| irf | in the file : ./etc/xinetd.d/tftp , user mentioned is : root | 11:22 |
| *** mgould has joined #openstack-ironic | 11:22 | |
| mgould | morning jrist | 11:23 |
| mgould | *jroll | 11:23 |
| jroll | hey mgould | 11:23 |
| dtantsur | jroll, looking at http://docs.openstack.org/releasenotes/ironic-python-agent/mitaka.html we've added a has_carrier flag to inventory, I wonder if we have to bump the minor version due to that | 11:25 |
| dtantsur | tonyb, ^^^ | 11:25 |
| jroll | dtantsur: idk, that's not an api we commit to | 11:26 |
| jroll | (afaik) | 11:26 |
| tonyb | dtantsur: do you have the review handy(ish) | 11:27 |
| dtantsur | tonyb, which review, introducing that change? | 11:27 |
| tonyb | dtantsur: yeah | 11:27 |
| dtantsur | tonyb, https://review.openstack.org/#/c/305916/ | 11:27 |
| jroll | vdrok: oh btw, is there a bug for this? | 11:28 |
| tonyb | dtantsur: Thanks. | 11:28 |
| vdrok | jroll: no, I can file one now | 11:28 |
| jroll | vdrok: thanks | 11:28 |
| dtantsur | tonyb, jroll, in theory, as we want to avoid adding features to a stable branch, we can hide this new attribute from API. it's like 1-line change.. just want to know what you guys think | 11:29 |
| dtantsur | FWIW I personally don't treat it as a real feature | 11:29 |
| *** stendulker has quit IRC | 11:30 | |
| mat128 | dtantsur: aren't new fields simply ignored by old clients? | 11:30 |
| dtantsur | they are | 11:30 |
| *** thrash|g0ne is now known as thrash | 11:31 | |
| dtantsur | I'm not saying it's a breaking change, I just checking with folks before I get problems with my release request :) | 11:31 |
| mat128 | oh, that was a backport | 11:31 |
| vdrok | jroll: https://bugs.launchpad.net/nova/+bug/1580987 | 11:31 |
| openstack | Launchpad bug 1580987 in OpenStack Compute (nova) ""Cannot call obj_load_attr on orphaned Instance object" in baremetal_basic_ops" [Undecided,New] | 11:31 |
| jroll | vdrok: thanks | 11:31 |
| jroll | dtantsur: the REST API doesn't expose the inventory, right? it's just what we submit to ironic | 11:32 |
| dtantsur | jroll, I think it's true | 11:32 |
| vdrok | brb | 11:32 |
| tonyb | dtantsur: sorry I was off reading the review so I don't sound like a moron ;P | 11:33 |
| tonyb | dtantsur: so there is a fine line between bug and feature at times. | 11:34 |
| tonyb | dtantsur: I think that this case is on the bug side of that line | 11:34 |
| *** sacharya has joined #openstack-ironic | 11:35 | |
| dtantsur | cool, thanks! then I'll submit the release request with 1.2.1 | 11:36 |
| tonyb | dtantsur: cool. | 11:36 |
| jroll | fyi, both of these patches independently (should) fix our gate issue, both are useful anyway :) https://review.openstack.org/#/c/315491/ and https://review.openstack.org/#/c/315495/ | 11:37 |
| jroll | vdrok: fyi | 11:37 |
| tonyb | dtantsur: others may disagree with me and I apologise if they do and it doubles your work | 11:37 |
| dtantsur | we'll see, no worries :) | 11:37 |
| dtantsur | oh, jroll, maybe we merge https://review.openstack.org/#/c/313123/ before the release? | 11:38 |
| *** milan_ has joined #openstack-ironic | 11:38 | |
| jroll | dtantsur: mehhhhhhhhhhhhhhhhhh | 11:38 |
| jroll | +A | 11:39 |
| dtantsur | ok, waiting for this to merge | 11:39 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: TEST Checking Gates https://review.openstack.org/311109 | 11:39 |
| *** sacharya has quit IRC | 11:40 | |
| *** baoli has joined #openstack-ironic | 11:46 | |
| *** baoli has quit IRC | 11:46 | |
| *** baoli has joined #openstack-ironic | 11:47 | |
| *** mjura has joined #openstack-ironic | 11:48 | |
| *** baoli_ has joined #openstack-ironic | 11:49 | |
| *** baoli has quit IRC | 11:52 | |
| sambetts | jroll: If your happy with approaching fixing our API URL collisions this way, then I'll get the UTs and the rest of the code written for https://review.openstack.org/#/c/314514/ | 11:53 |
| *** hamzy has joined #openstack-ironic | 11:54 | |
| *** wajdi has joined #openstack-ironic | 11:54 | |
| *** wajdi_ has joined #openstack-ironic | 11:56 | |
| *** absubram has quit IRC | 11:56 | |
| jroll | sambetts: so, this makes node.name=ports work right, yes? | 11:58 |
| sambetts | jroll: yup | 11:58 |
| jroll | nice | 11:58 |
| jroll | I like the code, cleans things up | 11:59 |
| *** ohamada has quit IRC | 11:59 | |
| jroll | wfm | 11:59 |
| *** dprince has joined #openstack-ironic | 11:59 | |
| *** ohamada has joined #openstack-ironic | 11:59 | |
| *** wajdi has quit IRC | 11:59 | |
| sambetts | cool :D I'll work out the UTs etc, and get a new patch set up | 11:59 |
| jroll | awesome, thanks for working on this | 12:00 |
| *** alexpilotti has quit IRC | 12:01 | |
| *** irf has quit IRC | 12:01 | |
| *** fragatina has joined #openstack-ironic | 12:01 | |
| *** alexpilotti has joined #openstack-ironic | 12:03 | |
| lucasagomes | sambetts, ++ | 12:03 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Test post don't upvote https://review.openstack.org/311865 | 12:07 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Test post don't upvote https://review.openstack.org/311865 | 12:08 |
| jroll | whee | 12:08 |
| mat128 | jroll: is that a test to find out automated bots? :) | 12:09 |
| jroll | hehe | 12:09 |
| jroll | still hacking on grenade things | 12:09 |
| jroll | it now actually has real code, but will need to be cleaned up | 12:09 |
| *** raildo-afk is now known as raildo | 12:10 | |
| lucasagomes | jroll, saw ur email from yesterday | 12:17 |
| lucasagomes | interesting | 12:17 |
| jroll | lucasagomes: right? sean and I just hacked out a thing | 12:18 |
| jroll | which reminds me I forgot depends-on | 12:18 |
| openstackgerrit | Jim Rollenhagen proposed openstack/ironic: Test post don't upvote https://review.openstack.org/311865 | 12:19 |
| jroll | here's the grenade change: https://review.openstack.org/#/c/315501/ | 12:19 |
| * lucasagomes looks | 12:19 | |
| jroll | and then 311865 there has us using it, needs some cleanup | 12:20 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-specs: Boot management for in-band inspection https://review.openstack.org/305864 | 12:21 |
| lucasagomes | jroll, good stuff, so the "early" (I kinda feel like it should be "pre") step does create a single flat tenant netwowrk | 12:23 |
| sambetts | jroll: did you see my email about making devstack using a flat network? | 12:25 |
| sambetts | jroll: I didn't know if that would just make everything work without having to change grenade, and it makes our devstack closer to a real deployment too | 12:28 |
| jroll | sambetts: yeah, something to investigate but that's a pretty huge change | 12:29 |
| jroll | lucasagomes: yep, we used 'early' because it does create things, it isn't before creating things | 12:29 |
| lucasagomes | hmm yeah makes sense | 12:30 |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic-specs: Boot management for in-band inspection https://review.openstack.org/305864 | 12:33 |
| *** pal_bth is now known as parlos | 12:34 | |
| dtantsur | jroll, tonyb, IPA release request: https://review.openstack.org/315520 | 12:36 |
| *** irf has joined #openstack-ironic | 12:36 | |
| jroll | dtantsur: +1'd | 12:39 |
| *** wajdi_ has quit IRC | 12:39 | |
| *** afaranha has joined #openstack-ironic | 12:41 | |
| *** athomas has quit IRC | 12:50 | |
| *** ohamada has quit IRC | 12:53 | |
| *** athomas has joined #openstack-ironic | 12:57 | |
| *** z- has joined #openstack-ironic | 12:58 | |
| *** [1]cdearborn has joined #openstack-ironic | 13:00 | |
| *** pcaruana|afk| is now known as pcaruana | 13:01 | |
| *** [1]cdearborn has quit IRC | 13:02 | |
| *** alexpilotti has quit IRC | 13:02 | |
| *** alexpilotti has joined #openstack-ironic | 13:03 | |
| *** absubram has joined #openstack-ironic | 13:04 | |
| *** irf has quit IRC | 13:06 | |
| *** aparnav has quit IRC | 13:07 | |
| *** keedya has joined #openstack-ironic | 13:07 | |
| *** rloo has joined #openstack-ironic | 13:14 | |
| *** links has quit IRC | 13:14 | |
| NobodyCam | Good Morning Ironicers | 13:15 |
| mgould | morning NobodyCam | 13:15 |
| NobodyCam | Morning mgould | 13:15 |
| NobodyCam | :) | 13:15 |
| NobodyCam | almost friday | 13:15 |
| NobodyCam | and the day I have to do my expense report :p | 13:18 |
| mgould | NobodyCam, d'oh | 13:19 |
| mgould | on the upside, it means your expenses will soon be paid? | 13:19 |
| * mgould only logged his expenses for last month's trip to Brno on Monday | 13:20 | |
| NobodyCam | lol +++ | 13:20 |
| *** amotoki has joined #openstack-ironic | 13:21 | |
| parlos | Good bye, and have a nice evening | 13:23 |
| mgould | parlos, good night! | 13:23 |
| thiagop | good morning Ironic | 13:25 |
| *** rbudden has joined #openstack-ironic | 13:25 | |
| mgould | morning thiagop | 13:25 |
| *** Guest51926 is now known as dansmith | 13:26 | |
| thiagop | mgould: o/ | 13:26 |
| NobodyCam | night parlos | 13:26 |
| NobodyCam | morning thiagop | 13:26 |
| thiagop | hiya NobodyCam | 13:27 |
| mjura | hi | 13:27 |
| lucasagomes | NobodyCam, hi there | 13:27 |
| mgould | hi mjura | 13:27 |
| *** Goneri has joined #openstack-ironic | 13:27 | |
| lucasagomes | thiagop, morning :-) | 13:27 |
| NobodyCam | hey good morning lucasagomes :) | 13:27 |
| *** parlos has quit IRC | 13:27 | |
| thiagop | lucasagomes: o/ | 13:27 |
| NobodyCam | morning mjura | 13:27 |
| sambetts | hey NobodyCam | 13:28 |
| mjura | I wanted to apply patch to ironic_lib for disk_utils on block_uuid | 13:28 |
| mjura | https://bugs.launchpad.net/ironic/+bug/1580989 | 13:28 |
| openstack | Launchpad bug 1580989 in Ironic "Couldn't determine the UUID of the root p artition or the disk identifier" [Undecided,New] | 13:28 |
| NobodyCam | morning sambetts :) | 13:28 |
| mjura | on my system it returns no UUID only PARTUUD | 13:28 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Prevent URL collisions with sub-controllers: nodes/ports https://review.openstack.org/314514 | 13:29 |
| mjura | I was wondering about putting there some condition to check if UUID is not null | 13:29 |
| mjura | if yes then get PARTUUID | 13:29 |
| mjura | I don't know if this is right solution | 13:29 |
| lucasagomes | mjura, which filesystem the image you are trying to deploy use? | 13:30 |
| mjura | lucasagomes: it is good question let me check it | 13:31 |
| mjura | lucasagomes: ext3 | 13:32 |
| lucasagomes | mjura, and the partition table? GPT? | 13:32 |
| mjura | lucasagomes: no normal dos partition | 13:32 |
| mjura | table | 13:32 |
| *** fragatina has quit IRC | 13:32 | |
| lucasagomes | hmm strange, AFAICT ext3 should have a uuid | 13:33 |
| lucasagomes | mjura, cause PARTUUID is a GPT thing, no? | 13:34 |
| *** wajdi has joined #openstack-ironic | 13:34 | |
| lucasagomes | it's two diff things the UUID we want there is the UUID from the filesystem of the image that was deployed | 13:34 |
| mjura | lucasagomes: after node is create it seems to have UUID and PARTUUID | 13:34 |
| mjura | but after dd UUID seems to not exist | 13:34 |
| lucasagomes | PARTUUID is the UUID of the GPT partition (AFAICT) | 13:34 |
| lucasagomes | mjura, the image you are trying to deploy is a whole disk image or a partition image? | 13:35 |
| lucasagomes | by whole disk image I mean, is it a image that already contains a partition table? | 13:35 |
| *** sacharya has joined #openstack-ironic | 13:36 | |
| mjura | lucasagomes: this is good question, but I assume that if UUID didn't exist so my images doesn't contain partition table | 13:37 |
| lucasagomes | mjura, I'm not sure, please check the image | 13:37 |
| *** joprovost has joined #openstack-ironic | 13:38 | |
| lucasagomes | mjura, because if it contains a partition table and Ironic tries to deploy it as a partition image | 13:38 |
| lucasagomes | it's very likely that it will fail to get the UUID of the filesystem | 13:38 |
| mjura | lucasagomes: ok, I'm checking image | 13:39 |
| lucasagomes | ack | 13:39 |
| mjura | lucasagomes: ok, I was wrong it has already dos partition table | 13:40 |
| lucasagomes | a-ha | 13:41 |
| lucasagomes | ok that explains it | 13:41 |
| *** fragatina has joined #openstack-ironic | 13:41 | |
| lucasagomes | so ironic is copying that whole image onto another partition | 13:41 |
| *** sacharya has quit IRC | 13:41 | |
| lucasagomes | and blkid won't find the filesystem identifiers because the beggining of the partition is pointing to the mbr from ur image | 13:41 |
| lucasagomes | mjura, you still can deploy that with ironic | 13:42 |
| lucasagomes | will find a link for u | 13:42 |
| *** dtantsur is now known as dtantsur|bbl | 13:42 | |
| mjura | lucasagomes: so it means that if I deploy my image as whole_disk_image | 13:42 |
| mjura | it should works | 13:42 |
| lucasagomes | exactly | 13:42 |
| mjura | ok, then I will check it once again | 13:42 |
| lucasagomes | mjura, http://docs.openstack.org/developer/ironic/deploy/install-guide.html#image-requirements see step 2) | 13:43 |
| lucasagomes | you just need to remove the kernel/ramdisk from the glance properties | 13:43 |
| sambetts | jroll, JayF: :( we need this too https://review.openstack.org/#/c/315536/ | 13:43 |
| mjura | ok, and then it will be booted also from local disk | 13:43 |
| mjura | using mbr from my image | 13:44 |
| *** joprovost has quit IRC | 13:44 | |
| lucasagomes | mjura, precisely | 13:44 |
| mjura | lucasagomes: awesome, now I understand it :) | 13:44 |
| lucasagomes | :-) | 13:45 |
| mjura | is there any easy way to convert partiotioned image to non-partitioned | 13:45 |
| jroll | sambetts: heh, did we blow away the master build? | 13:45 |
| *** joprovost has joined #openstack-ironic | 13:46 | |
| lucasagomes | mjura, kinda-ish... if you have only 1 partition that's easy ebcause you can mount it in a loop device | 13:46 |
| lucasagomes | and dd the partition onto another file | 13:46 |
| sambetts | jroll: no luckly, the build failed because it couldn't find tinyipa-mitaka to upload | 13:46 |
| jroll | ah | 13:46 |
| mjura | ok, and in this way I will get raw image | 13:46 |
| lucasagomes | mjura, right | 13:46 |
| jroll | sambetts: +2 | 13:46 |
| lucasagomes | mjura, you can convert to qcow2 later with qemu-img | 13:47 |
| mjura | yes, sure | 13:47 |
| wajdi | good day fellow metalers | 13:48 |
| mjura | lucasagomes: then I will invalid this bug and I will add explanation to it | 13:48 |
| lucasagomes | if you have more than 1 partition I would usggest u use libguestfs after copying the root partition to edit the mtab | 13:48 |
| *** joprovost has quit IRC | 13:48 | |
| lucasagomes | mjura, thanks much! | 13:48 |
| sambetts | jroll: cool :) I've also uploaded backports for the tinyIPA optimisations if we're allowed to do | 13:48 |
| mjura | lucasagomes: thank you very much, you helped me a lot :) | 13:48 |
| lucasagomes | wajdi, hi there | 13:48 |
| wajdi | hey :) | 13:49 |
| lucasagomes | mjura, you are very welcome! | 13:49 |
| mat128 | o/ wajdi ;) | 13:49 |
| wajdi | :) hey mat | 13:49 |
| mjura | big thank you once again :) yupi | 13:49 |
| lucasagomes | yw :-) | 13:50 |
| *** joprovost has joined #openstack-ironic | 13:50 | |
| jroll | sambetts: yeah, I think that's a good thing but I might bug stable folks about it | 13:50 |
| sambetts | :) cool | 13:51 |
| openstackgerrit | Milan Kováčik proposed openstack/ironic-inspector: [WIP] Tempest: add basic test https://review.openstack.org/293926 | 13:54 |
| jroll | as a note, neither of my nova patches fixed the issue in the gate :( | 13:54 |
| mgould | morning wajdi mat128 | 13:55 |
| wajdi | o/ | 13:55 |
| *** cdearborn has joined #openstack-ironic | 13:57 | |
| openstackgerrit | Andre keedy proposed openstack/ironic-inspector: Replace node data via uuid/data/unprocessed API https://review.openstack.org/308379 | 13:58 |
| NobodyCam | morning jroll | 13:58 |
| NobodyCam | :) | 13:58 |
| *** fragatina has quit IRC | 13:59 | |
| mat128 | hello mgould | 13:59 |
| *** penick has quit IRC | 14:00 | |
| *** joprovost has quit IRC | 14:02 | |
| *** joprovost1 has joined #openstack-ironic | 14:02 | |
| *** joprovost1 is now known as joprovost | 14:04 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Raise an exception for domains already registered https://review.openstack.org/315149 | 14:05 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Add unittests for the vbmc.py module https://review.openstack.org/315589 | 14:05 |
| * lucasagomes coverage up to 95% now | 14:05 | |
| NobodyCam | lucasagomes: awesome ! | 14:05 |
| sambetts | lucasagomes: sweet! | 14:05 |
| lucasagomes | yeah, now I will put some jobs in the vbmc gate | 14:05 |
| lucasagomes | we already have one in pyghmi | 14:05 |
| lucasagomes | and I think that will guard any breaking changes to the ironic jobs | 14:05 |
| lucasagomes | I've also put a patch to bump the version of pyghmi in the g-r since some mid versions had a regression | 14:06 |
| keedya | GMorning 0/ | 14:08 |
| mat128 | lucasagomes: woot! | 14:10 |
| NobodyCam | morning keedya | 14:11 |
| *** ChrisAusten has joined #openstack-ironic | 14:12 | |
| *** d0ugal has quit IRC | 14:14 | |
| *** rloo has quit IRC | 14:14 | |
| *** causten_ has joined #openstack-ironic | 14:15 | |
| *** rloo has joined #openstack-ironic | 14:15 | |
| *** xavierr has quit IRC | 14:16 | |
| *** d0ugal has joined #openstack-ironic | 14:16 | |
| *** ChrisAusten has quit IRC | 14:17 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: DevStack: Support to install virtualbmc from source https://review.openstack.org/315593 | 14:19 |
| *** joprovost1 has joined #openstack-ironic | 14:22 | |
| *** d0ugal has quit IRC | 14:23 | |
| *** joprovost has quit IRC | 14:23 | |
| *** joprovost1 is now known as joprovost | 14:23 | |
| rloo | morning everyone, lucasagomes, NobodyCam, mat128, keedya, sambetts, wajdi, mgould, jroll | 14:34 |
| lucasagomes | rloo, hi there, morning | 14:35 |
| openstackgerrit | Hugo Nicodemos proposed openstack/python-oneviewclient: Adding dynamic allocation operations https://review.openstack.org/258699 | 14:35 |
| rloo | jroll, do/should we agree on whether or not to split up the network providers patch? Someone is asking and I don't want to review until I know: https://review.openstack.org/285852 | 14:35 |
| wajdi | morning rloo :) | 14:36 |
| *** links has joined #openstack-ironic | 14:36 | |
| rloo | sambetts: wrt that network providers patch ^^, didn't you say there was a problem (after summit core party i think) with the dhcp_provider? I see your latest comment but was that the problem? | 14:38 |
| *** mjura has quit IRC | 14:40 | |
| *** thrash is now known as thrash|afk | 14:40 | |
| sambetts | rloo: yeah, I've pointed it out in the migration, but I'm not sure if handling it there is the right solution or not | 14:41 |
| rloo | sambetts: ok, that's what i was wondering. thx. | 14:41 |
| *** stendulker has joined #openstack-ironic | 14:43 | |
| *** joprovost has quit IRC | 14:44 | |
| jroll | rloo: what do you think? don't let me dictate :) | 14:44 |
| jroll | rloo: what I had told people is that it will be easier to review / get reviews / land if it's broken up | 14:45 |
| rloo | jroll: i think it should be split up. I thought it should be split up but at the time, also thought we were close to getting it all in and it was just me that thought it should be split up. Now that I know that others think the same thing... | 14:45 |
| jroll | rloo: yeah, agree | 14:45 |
| jroll | rloo: I wonder if one of us cores should do that | 14:46 |
| rloo | fwiw, splitting it up now will actually make it harder for me to review since i may need to compare before/after. but i think it is the right thing to do. | 14:46 |
| rloo | jroll: also, if we split it up, could we put these BEFORE the portgroup API patch? | 14:46 |
| jroll | rloo: maybe? probably? | 14:46 |
| rloo | jroll: i could be wrong, but i don't think it depends on the portgroups one. | 14:47 |
| rloo | jroll: and if we put it before, could we land some of these sooner, or would it all have to wait til we have grenade working? | 14:47 |
| jroll | rloo: I'd like upgrade testing on the driver load refactoring | 14:47 |
| jroll | that's the big thing for me | 14:47 |
| rloo | jroll: my feeling is why review now when it might be easier/faster to review later when we have testing. | 14:48 |
| jroll | rloo: sure | 14:48 |
| rloo | jroll: yeah, that makes sense. so you answered my question. | 14:48 |
| rloo | jroll: it would be more 'efficient' for us to review *after* we have testing in place. | 14:48 |
| sambetts | ++ | 14:48 |
| * sambetts wonders if some upgrade senarios won't be caught by grenade e.g. the dhcp_providers = None problem | 14:49 | |
| rloo | if we/I am going to wait til after we have testing in place, then 1. still worth splitting up the network providers patch; 2. not sure it matters if it is before/after portgroup API patch | 14:50 |
| rloo | sambetts: very difficult for tests to test ALL scenarios | 14:50 |
| sambetts | yeah :/ | 14:50 |
| sambetts | EEETOOMANYCONFIGPURMUTATIONS | 14:50 |
| NobodyCam | morning rloo :) | 14:51 |
| NobodyCam | and expense report done | 14:52 |
| lazy_prince | Hi.. is there any plan to backport c6e9bfd6b1de8c7bd8103a0c58ca51bf8e56e85c to stable/mitaka for python-ironicclient..? | 14:52 |
| rloo | hi vsaienko: do you have time and would you be willing to split up https://review.openstack.org/#/c/285852/ into smaller patches? | 14:55 |
| mgould | rloo, could we use fuzz-testing? | 14:55 |
| rloo | mgould: dunno. sounds too fuzzy for me. what is it? | 14:56 |
| mgould | random search of config-space, testing for properties that should always hold like "does not crash" | 14:56 |
| mgould | ideally with some kind of reducer to find minimal breaking test cases | 14:56 |
| rloo | mgould: OH. I was thinking you were referring to the splitting up of a patch. | 14:57 |
| mgould | that would also be cool | 14:57 |
| rloo | mgould: you asking that as a 'general' test? | 14:57 |
| mgould | yeah, I guess so | 14:57 |
| mgould | in response to sambetts "EEETOOMANYCONFIGPURMUTATIONS" | 14:57 |
| rloo | mgould: in theory, i would say why not. more testing the better. | 14:58 |
| *** rama_y has joined #openstack-ironic | 14:58 | |
| mgould | Python fuzz-testing library: http://hypothesis.readthedocs.io/en/latest/ | 14:58 |
| mgould | rloo, awesome | 14:58 |
| rloo | mgould: OH. well, i think the problem there is the amount of time/resources it would take to run some of our tests. | 14:58 |
| rloo | mgould: but it would be interesting. | 14:58 |
| mgould | I'll look for opportunities to try it out | 14:58 |
| mgould | fuzzing the whole process end-to-end would be a big problem, definitely | 14:58 |
| mgould | but I'll look for places where we can use it for unit testing | 14:59 |
| mat128 | mgould: fuzzy-testing end-to-end tempest tests is asking for trouble :) | 15:00 |
| mat128 | and unreproductible runs | 15:00 |
| mat128 | but for unit tests I definitely see it working | 15:00 |
| mgould | yeah, I don't know what best practice is there | 15:01 |
| mat128 | I'll try it here on some project :) | 15:01 |
| mgould | :-) | 15:01 |
| JayF | https://review.openstack.org/#/c/313035 is landable now, the project-config change it depended on merged and was puppet applied overnight | 15:01 |
| JayF | I'll send an email about it during this meeting | 15:01 |
| mgould | I think you can save the random seed and/or record inputs that cause failures, but our gate's already quite nondeterministic | 15:01 |
| vsaienko | hi rloo, I will try to do it tomorrow. At the moment I'm working on grenade job | 15:02 |
| rloo | thanks vsaienko. No hurry, next week is fine. I just want to make sure we had a plan :) | 15:03 |
| *** sivaramakrishna has joined #openstack-ironic | 15:04 | |
| vdrok | morning rloo and NobodyCam :) | 15:04 |
| vdrok | lazy_prince: hey, forgot about that, thanks for reminder! will propose a backport now | 15:05 |
| openstackgerrit | Vasyl Saienko proposed openstack/ironic: TEST Checking Gates https://review.openstack.org/311109 | 15:05 |
| lazy_prince | vdrok: yeah.. its nasty one and needs to be back-ported... | 15:06 |
| NobodyCam | morning vdrok | 15:06 |
| NobodyCam | vdrok: awesome TY | 15:07 |
| NobodyCam | I'm about to test it in my env | 15:07 |
| vdrok | jroll: dtantsur|bbl please take a look https://review.openstack.org/#/c/315623/1 | 15:09 |
| *** joprovost has joined #openstack-ironic | 15:15 | |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Nextscale total power capacity as a sensor https://review.openstack.org/315628 | 15:16 |
| *** ohamada has joined #openstack-ironic | 15:17 | |
| *** d0ugal has joined #openstack-ironic | 15:17 | |
| mat128 | lazy_prince, NobodyCam: wow, we've been hit by c6e9bfd6b1de8c7bd8103a0c58ca51bf8e56e85c here, thought it was something in keystoneclient | 15:19 |
| *** shakamunyi has quit IRC | 15:19 | |
| mat128 | it just took first region in catalog and called it a day :( | 15:19 |
| NobodyCam | mat128: ++ | 15:19 |
| Madasi | .buffer h-a | 15:19 |
| Madasi | and, misfire :( | 15:19 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add Nextscale total power capacity as a sensor https://review.openstack.org/315628 | 15:20 |
| lazy_prince | mat128: same here.. all this time I was thinking it as a keystone bug.. but turned out to be a ironicclient bug.. | 15:20 |
| mat128 | lazy_prince: we solved it by properly using upper-constraints.txt when building our packages, but that just avoided the bug | 15:21 |
| *** sacharya has joined #openstack-ironic | 15:21 | |
| * NobodyCam likes mat128 's what bug approach | 15:22 | |
| lazy_prince | mat128: ack. but its worth back-porting this to mitaka since so many are affected... | 15:22 |
| NobodyCam | +++ | 15:22 |
| mat128 | lazy_prince: definitely! | 15:22 |
| *** harlowja_at_home has joined #openstack-ironic | 15:22 | |
| jroll | lazy_prince: if you want something backported, just do it and bug us :) | 15:23 |
| lazy_prince | is there a way to prevent such things in future.. say as part of gate job, we check if internalURL works... or something similar..? | 15:24 |
| mat128 | lazy_prince: I think we would require multi-region support to test these things | 15:24 |
| lazy_prince | jroll: vdrok already backported it.. https://review.openstack.org/#/c/315623/1 | 15:24 |
| jroll | lazy_prince: ah, cool, thanks | 15:25 |
| JayF | mat128: sambetts: Did either of you look at why post job failed after last backport was done to help fix it/ | 15:25 |
| lazy_prince | mat128: not relly.. I hit this bug in a single region env.. I just tried using internalURL as I did not have publicURL in my env.. | 15:26 |
| *** sacharya_ has joined #openstack-ironic | 15:26 | |
| vdrok | lazy_prince: mat128 yep, I think we would have to create a 2 node job with ir-api and ir-cond on both in diff regions | 15:26 |
| JayF | mat128: sambetts: I can if needed, just didn't want to start from zero if work has been done | 15:26 |
| *** catintheroof has joined #openstack-ironic | 15:26 | |
| sambetts | JayF: yup, we need https://review.openstack.org/#/c/315536/ | 15:27 |
| *** catintheroof has quit IRC | 15:27 | |
| JayF | tonyb: still around? https://review.openstack.org/#/c/315536/ is needed too | 15:28 |
| *** parlos has joined #openstack-ironic | 15:28 | |
| JayF | sambetts: okie, I'm also presuming once all these optimizations get in and prove at lower ram in the gate, we'll need to backport all of those too, right? | 15:28 |
| *** sacharya has quit IRC | 15:28 | |
| jroll | update on gate breakage things: https://review.openstack.org/#/c/312751/10 fixes this, which was later in the chain for the thing that broke this, turns out it was a bad rebase | 15:28 |
| *** catintheroof has joined #openstack-ironic | 15:29 | |
| sambetts | JayF: I've got all the patches up to backport them, but its down to if stable let us do it, if not we'll have to add a conditional in the gate job, if mitaka then 384 elif >mitaka 256 | 15:29 |
| JayF | sambetts: for grenade, we have to be able to run in mitaka | 15:30 |
| JayF | sambetts: and we want to be able to run more ram for grenade | 15:30 |
| mat128 | JayF: tbh I havent looked at the post job, where do you see the job log other than watching zuul while it's running? | 15:30 |
| JayF | sambetts: which made me think we need a critical bug like "IPA requires too much ram to run many ramdisks in a job" | 15:30 |
| JayF | logs.openstack.org/[first two digits of sha]/[full sha]/ | 15:30 |
| sambetts | JayF: makes sense I guess, the patches are up to do it, they are in a chain with the patch I just sent you | 15:31 |
| JayF | for example, the breakage that was yesterday that I backported something to fix: http://logs.openstack.org/ed/ed978f312e1079c6eb7166947253007d141eb82d/ | 15:31 |
| *** hshiina has quit IRC | 15:31 | |
| mat128 | JayF: oh great :) | 15:31 |
| mat128 | I was expecting logs to appear in gerrit right below check/gate stuff | 15:32 |
| mat128 | s/logs/links to logs/ | 15:32 |
| *** links has quit IRC | 15:32 | |
| mat128 | made him disconnect ^ | 15:32 |
| mat128 | ;) | 15:32 |
| JayF | if your /nick is a common word, no getting upset for too many highlights | 15:33 |
| mat128 | yup | 15:34 |
| *** sacharya_ has quit IRC | 15:34 | |
| milan_ | guys, how does one run tempest? | 15:34 |
| mat128 | sambetts: are you sure it's only missing the -branch thing? in JayF's link (ending in 1eb82d), there's a "2016-05-11 21:41:56.686 | ./build-tinyipa.sh: line 85: mksquashfs: command not found" | 15:35 |
| mat128 | or was that fixed later? | 15:35 |
| *** dmk0202 has quit IRC | 15:35 | |
| sambetts | mat128: that was fixed by JayF's backport that landed this morning | 15:36 |
| rloo | JayF: you want this +A'd now? https://review.openstack.org/#/c/313035/ | 15:36 |
| *** dmk0202 has joined #openstack-ironic | 15:36 | |
| JayF | rloo: It can be +A'd safely now. As to if it should land, you're the core, not me :P | 15:36 |
| *** e0ne has quit IRC | 15:36 | |
| * JayF sets seatbacks to full and upright position and stows his tray table for landing | 15:36 | |
| rloo | JayF: 3 cores sez yes :) | 15:37 |
| mat128 | sambetts: found it, f5b7ea238bbf6702b33de67289020a590cfb6e28 thanks :) | 15:38 |
| *** mtanino has joined #openstack-ironic | 15:41 | |
| *** sacharya has joined #openstack-ironic | 15:43 | |
| vdrok | milan_: http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#running-tempest-tests | 15:44 |
| milan_ | vdrok, thx | 15:44 |
| * milan_ reads | 15:44 | |
| jroll | JayF: sambetts: I think 384 should be fine for grenade for now | 15:46 |
| openstackgerrit | Mathieu Mitchell proposed openstack/ironic-python-agent: Make TinyCore mirror configurable https://review.openstack.org/315649 | 15:49 |
| *** xhku has joined #openstack-ironic | 15:50 | |
| JayF | mat128: do you know if that can use https? | 15:50 |
| JayF | mat128: to pull down the tinycore files? | 15:50 |
| mat128 | JayF: checking | 15:50 |
| mat128 | was ibiblio https? | 15:50 |
| JayF | mat128: just occured to me how MITM our build can be, even in master, for tinyipa | 15:50 |
| JayF | no | 15:50 |
| JayF | but I saw this while reviewing that ;) | 15:50 |
| mat128 | also, requested mirror to -infra: http://lists.openstack.org/pipermail/openstack-infra/2016-May/004277.html | 15:51 |
| vdrok | jroll: jlvillal smoke tests almost pass :) http://logs.openstack.org/09/311109/7/experimental/gate-grenade-dsvm-ironic/8404ae5/logs/old/testr_results.html.gz one test seem to have failed because of OS_TEST_TIMEOUT | 15:51 |
| jlvillal | vdrok: Very cool :) | 15:51 |
| vdrok | jlvillal: now I say privyet :) | 15:52 |
| *** harlowja_at_home has quit IRC | 15:52 | |
| jroll | vdrok: I've had them passing since last week though? | 15:52 |
| mat128 | JayF: no https for either ibiblio or official url | 15:52 |
| mat128 | JayF: Checking other mirrors | 15:52 |
| jlvillal | vdrok: Privyet. Kak dela? | 15:52 |
| * jroll looks at what vdrok changes | 15:53 | |
| vdrok | jroll: hm, on the gate? | 15:53 |
| jroll | vdrok: yep | 15:53 |
| *** MattMan has quit IRC | 15:54 | |
| jroll | vdrok: oh apparently not? O_o | 15:54 |
| jroll | grenade just continues | 15:54 |
| jroll | http://logs.openstack.org/65/311865/4/experimental/gate-grenade-dsvm-partial-ironic/35420fe/logs/old/testr_results.html.gz | 15:54 |
| *** MattMan has joined #openstack-ironic | 15:54 | |
| jroll | wait | 15:54 |
| jroll | wrong log | 15:54 |
| jroll | vdrok: http://logs.openstack.org/65/311865/3/experimental/gate-grenade-dsvm-ironic/e635dec/logs/old/testr_results.html.gz | 15:55 |
| jroll | vdrok: pretty minimal tempest changes https://review.openstack.org/#/c/311864/ | 15:55 |
| mat128 | JayF: confirmed, none offer HTTPS, but http://www.gtlib.gatech.edu/pub/tinycore/ is much faster than the other ones. | 15:55 |
| JayF | mat128: whomp, I think it's "ok" for now, but we should probably do some gpg + md5 goodness there | 15:56 |
| mat128 | JayF: Maybe we can get away with it with verifying the checksum | 15:56 |
| JayF | mat128: well, that's all going to depend on location, right? | 15:56 |
| mat128 | great minds think alike :) | 15:56 |
| *** garthb has joined #openstack-ironic | 15:56 | |
| JayF | mat128: lets just merge your change as-is | 15:56 |
| JayF | and why don't you file a bug about it/ | 15:56 |
| JayF | so we don't forget but don't get rabbitholed here | 15:56 |
| mat128 | upstream? | 15:56 |
| mat128 | oh | 15:56 |
| JayF | of course | 15:56 |
| mat128 | you mean, in their forum/bug tracker? | 15:56 |
| mat128 | to offer https? | 15:56 |
| JayF | no, in ours | 15:56 |
| JayF | for us to not blindly accept any randomass pacakge given to us, and to check gpg/md5, etc | 15:57 |
| mat128 | JayF: writing it atm | 15:57 |
| JayF | cool | 15:57 |
| vdrok | jroll: heh, seems like it :) this one https://review.openstack.org/#/c/315422/ should be able to do it on 3 vms | 15:58 |
| *** jcoufal has quit IRC | 16:00 | |
| mat128 | JayF: Shall I check the "security issue" checkbox? I mean, us discussing it means it's public already, isn't it? | 16:00 |
| JayF | I wouldn't check that checkbox. | 16:01 |
| *** alexpilotti has quit IRC | 16:01 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Add unittests for the vbmc.py module https://review.openstack.org/315589 | 16:02 |
| *** alexpilotti has joined #openstack-ironic | 16:02 | |
| *** jtomasek has quit IRC | 16:03 | |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/virtualbmc: Add unittests for the vbmc.py module https://review.openstack.org/315589 | 16:04 |
| mat128 | JayF: https://bugs.launchpad.net/ironic-python-agent/+bug/1581112 | 16:04 |
| openstack | Launchpad bug 1581112 in ironic-python-agent "TinyIPA build is subject to MITM" [Undecided,New] | 16:04 |
| jroll | vdrok: ah, I see | 16:06 |
| openstackgerrit | Merged openstack/virtualbmc: Raise an exception for domains already registered https://review.openstack.org/315149 | 16:14 |
| *** dmk0202 has quit IRC | 16:15 | |
| *** chopmann has quit IRC | 16:17 | |
| *** milan_ has quit IRC | 16:19 | |
| *** d0ugal has quit IRC | 16:20 | |
| *** d0ugal has joined #openstack-ironic | 16:21 | |
| lucasagomes | is the gate broken? | 16:24 |
| *** lintan has quit IRC | 16:25 | |
| JayF | jroll: was saying something to that effect, I think | 16:25 |
| JayF | something in nova/ | 16:25 |
| jroll | lucasagomes: yes | 16:25 |
| lucasagomes | :-( | 16:25 |
| * lucasagomes checks whiteboard | 16:25 | |
| jroll | 15:28:51 jroll | update on gate breakage things: https://review.openstack.org/#/c/312751/10 fixes this, which was later in the chain for the thing that broke this, turns out it was a bad rebase | 16:25 |
| *** trusted has joined #openstack-ironic | 16:25 | |
| *** rm_work has joined #openstack-ironic | 16:26 | |
| lucasagomes | jroll, right, sorry wasn't paying attention tot he channel | 16:26 |
| * lucasagomes updates the whiteboard | 16:26 | |
| jroll | no worries | 16:26 |
| jroll | I just did, I think | 16:26 |
| lucasagomes | Gate Status: GOOD? NO heh | 16:26 |
| jroll | :) | 16:27 |
| *** d0ugal has quit IRC | 16:27 | |
| *** fragatina has joined #openstack-ironic | 16:27 | |
| *** sivaramakrishna has quit IRC | 16:28 | |
| *** lintan has joined #openstack-ironic | 16:29 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: devstack: Fetch tarball images via https https://review.openstack.org/315663 | 16:31 |
| * jlvillal will see if anyone objects to using https instead of http | 16:31 | |
| lucasagomes | jlvillal, hope nobody! | 16:32 |
| *** trusted has quit IRC | 16:36 | |
| * sambetts continues to work on refactors the API controllers, might end up with PTSD | 16:38 | |
| jroll | lol | 16:38 |
| jlvillal | lucasagomes: :) | 16:40 |
| *** kromanenko_ has quit IRC | 16:41 | |
| *** d0ugal has joined #openstack-ironic | 16:42 | |
| *** ChrisAusten has joined #openstack-ironic | 16:46 | |
| *** causten_ has quit IRC | 16:46 | |
| *** ChrisAusten has quit IRC | 16:47 | |
| *** ChrisAusten has joined #openstack-ironic | 16:47 | |
| *** alex_xu has quit IRC | 16:49 | |
| *** heyongli has quit IRC | 16:50 | |
| *** stendulker has quit IRC | 16:50 | |
| *** heyongli has joined #openstack-ironic | 16:52 | |
| *** alex_xu has joined #openstack-ironic | 16:52 | |
| openstackgerrit | Merged openstack/virtualbmc: Add unittests for the vbmc.py module https://review.openstack.org/315589 | 16:53 |
| *** garthb_ has joined #openstack-ironic | 16:53 | |
| lucasagomes | folks I'm calling it a day | 16:54 |
| lucasagomes | have a great evening all, see you all tomorrow | 16:54 |
| * sambetts is also heading off | 16:54 | |
| sambetts | cya lucasagomes | 16:54 |
| NobodyCam | have a good night lucasagomes | 16:54 |
| lucasagomes | sambetts, see ya! | 16:54 |
| sambetts | night all | 16:54 |
| *** lucasagomes is now known as lucas-dinner | 16:54 | |
| NobodyCam | night sambetts | 16:54 |
| *** sambetts is now known as sambetts|afk | 16:55 | |
| wajdi | goodnight lucas-dinner sambetts|afk | 16:55 |
| *** garthb has quit IRC | 16:55 | |
| *** elfosardo has quit IRC | 16:56 | |
| rm_work | ah I missed sambetts, lol... anyone know what timezone he's in? | 16:58 |
| rm_work | UK? | 16:58 |
| JayF | yes | 16:58 |
| rm_work | right, I think someone did mention that earlier. alright, tomorrow then :P | 16:59 |
| *** lintan_ has joined #openstack-ironic | 17:01 | |
| *** lintan has quit IRC | 17:02 | |
| *** e0ne has joined #openstack-ironic | 17:03 | |
| *** ijw has joined #openstack-ironic | 17:05 | |
| *** nicodemos has quit IRC | 17:06 | |
| *** thrash|afk is now known as thrash | 17:06 | |
| * mgould -> home; good night! | 17:09 | |
| *** ChubYann has joined #openstack-ironic | 17:09 | |
| *** ijw has quit IRC | 17:10 | |
| wajdi | o/ mgould | 17:10 |
| NobodyCam | night mgould | 17:11 |
| *** ijw has joined #openstack-ironic | 17:12 | |
| mgould | wajdi, NobodyCam goodnight! | 17:14 |
| *** mgould has quit IRC | 17:14 | |
| NobodyCam | :) | 17:14 |
| *** davideagnello has joined #openstack-ironic | 17:14 | |
| afaranha | Hey guys, since yesterday I've been trying to test the integration code but with no success. Is this local.conf here https://review.openstack.org/#/c/258596/64/doc/source/dev/ironic-neutron-integration.rst updated? The line 39 "enable_plugin networking-generic-switch https://review.openstack.org/openstack/networking-generic-switch" is correct? | 17:21 |
| *** dtantsur|bbl is now known as dtantsur | 17:24 | |
| *** jistr has quit IRC | 17:24 | |
| *** e0ne has quit IRC | 17:24 | |
| dtantsur | morning NobodyCam, jlvillal, rloo and everyone I've missed while was out :) | 17:24 |
| jlvillal | dtantsur: Welcome back :) | 17:24 |
| rloo | hi dtantsur, isn't it late-ish for you? | 17:24 |
| NobodyCam | morning dtantsur and rloo :) | 17:25 |
| rloo | hi NobodyCam | 17:25 |
| NobodyCam | :) | 17:25 |
| dtantsur | rloo, well, it is :) I was out for some time, so I'll stay around for some more time | 17:25 |
| *** garthb__ has joined #openstack-ironic | 17:25 | |
| dtantsur | 7:25pm now | 17:25 |
| NobodyCam | lol | 17:25 |
| rloo | dtantsur: nothing like chillaxing in the evening with a glass of wine and good code :D | 17:26 |
| *** garthb_ has quit IRC | 17:26 | |
| dtantsur | hehe | 17:26 |
| devananda | dtantsur: as inspector gathers the Port MAC addresses from a node, do you have any thoughts on how it might record the system device order? | 17:30 |
| dtantsur | devananda, not sure I get it.. what's the use case? | 17:31 |
| rloo | dtantsur (and devananda if he has an opinion cuz he reviewed): was there a vote or decision on this: see line 128ish: https://review.openstack.org/#/c/275766/12/ironic/api/controllers/v1/node.py | 17:33 |
| rloo | dtantsur: cuz I agree with you | 17:33 |
| devananda | greghaynes could elaborate more. tldr; an environment we're working in seems to set up NIC bonding based on system device order | 17:34 |
| dtantsur | rloo, I don't remember voting on it.. my position is kind of the same, though I'm not too religious about it | 17:34 |
| dtantsur | devananda, does is match with the name order? | 17:34 |
| *** d0ugal has quit IRC | 17:34 | |
| rloo | dtantsur: it seems wrong. adopt is NOT deploy | 17:34 |
| devananda | dtantsur: purely by chance | 17:34 |
| dtantsur | aha | 17:35 |
| dtantsur | devananda, so how does one figure out this order? | 17:35 |
| rloo | dtantsur: I'll comment... | 17:35 |
| devananda | dtantsur: walk /sys during inspection | 17:35 |
| dtantsur | devananda, but that's name order, no? could you show me a bash snippet, maybe I'm just stupid in the evening :) | 17:35 |
| devananda | dtantsur: heh, seems like we found another way to do this, so we don't need to store the device order right now after all | 17:36 |
| *** ChrisAusten has quit IRC | 17:36 | |
| dtantsur | ok :) | 17:36 |
| dtantsur | but would be interesting to know anyway | 17:36 |
| *** alex_xu has quit IRC | 17:38 | |
| NobodyCam | morning devananda :) | 17:38 |
| *** Sukhdev has joined #openstack-ironic | 17:42 | |
| * jroll finishes three hours of meetings and finds sustenance | 17:42 | |
| *** e0ne has joined #openstack-ironic | 17:44 | |
| mat128 | Hey guys, I requested (yeah, I know) a Tiny Core mirror on the openstack-infra ML and was told to do my homework :), given I figure out how to properly mirror Tiny Core, where would I add this? (c.f. "interested stakeholders did the work") | 17:50 |
| mat128 | I'm trying to find out where we are building our CentOS or Ubuntu mirrors and can't find anything | 17:50 |
| mat128 | http://lists.openstack.org/pipermail/openstack-infra/2016-May/004278.html | 17:50 |
| JayF | ask in -infra | 17:51 |
| JayF | but I think they use bandersnatch | 17:51 |
| JayF | maybe it's in system-config? | 17:51 |
| JayF | https://github.com/openstack-infra/puppet-bandersnatch | 17:51 |
| JayF | idk what calls it though | 17:52 |
| *** d0ugal has joined #openstack-ironic | 17:52 | |
| mat128 | hmm, that'd be for pip | 17:52 |
| mat128 | I,ll look it up in the various -infra repos | 17:52 |
| JayF | mat128: I'm asking in -infra now ;) | 17:53 |
| mat128 | yeah, got mentioned | 17:53 |
| cinerama | hi JayF | 17:54 |
| JayF | hey | 17:54 |
| openstackgerrit | Merged openstack/pyghmi: Add Nextscale total power capacity as a sensor https://review.openstack.org/315628 | 17:55 |
| cinerama | so re https://review.openstack.org/#/c/315213/, it's a little complicated because we don't actually pip install ansible by default. we clone it into a directory and the user runs their hacking env setup script to do appropriate environment things to run ansible from the checkout | 17:55 |
| JayF | jroll: does Ironic have an infra liason? | 17:55 |
| JayF | jroll: if not, I'll gladly signup for it | 17:55 |
| cinerama | we *could* switch to always pip installing ansible but that's a change to our behaviour that i don't want to do right now just to get the gate going | 17:56 |
| JayF | I mean, I'm extremely -1 to adding a non-requirement to the requirements list. | 17:57 |
| JayF | Cores can feel free merge it over my -1 if they want, but I think you're just pushing the bigger problem out a little further | 17:57 |
| JayF | the gate broke because something is actually broken | 17:57 |
| JayF | and instead of fixing it, we're working around it, which doesn't seem like the right thing to do | 17:58 |
| cinerama | i realize it's not ideal, and it's something we'd like to fix, but we have precedent for maintaining ansible's deps there. | 17:58 |
| JayF | You already have some ansible deps defined there? | 17:58 |
| *** pcaruana has quit IRC | 17:58 | |
| JayF | gah | 17:58 |
| NobodyCam | JayF: ya we do | 17:59 |
| cinerama | ansible don't have a requirements file either | 17:59 |
| NobodyCam | I agree that we need to rethink how we are addressing things | 17:59 |
| cinerama | (otherwise we could pip install stuff from there in our setup) | 17:59 |
| JayF | Is there a bug somewhere about making that more sane? If we've already started down the path one more doesn't make it worse | 17:59 |
| cinerama | JayF: that's the exact plan | 17:59 |
| JayF | but we should at least make sure we have the bug filed that says "this is wrong, we know it, and we should fix it" | 17:59 |
| NobodyCam | but I was hoping to do that once TheJulia returned from PTO | 17:59 |
| JayF | just make https://bugs.launchpad.net/bifrost/+bug/1580777 that bug | 18:00 |
| openstack | Launchpad bug 1580777 in bifrost "Gate is broken due to missing pycrypto dependency" [Critical,New] - Assigned to Stephane Miller (stephaneeee) | 18:00 |
| NobodyCam | JayF: ++ to bug! | 18:00 |
| JayF | and mark the gate fix as only related to it | 18:00 |
| JayF | because the real bug is that we statically list ansible deps vs anyhting else | 18:00 |
| cinerama | yeah we do not want to significantly change our behaviour without consulting with julia | 18:00 |
| JayF | Well you all might wanna try and fix that too :P I've been a single point of failure before, it's a lot of pressure, I doubt julia wants to be one either | 18:01 |
| JayF | haha | 18:01 |
| *** ohamada has quit IRC | 18:01 | |
| cinerama | i will revise that bug to reflect root cause rather than symptoms, and revise the patch to make it "partial bug" | 18:01 |
| JayF | but I get it, and if you've already crossed the bridge of having ansible deps in there my objection stops being useful and starts being pedantic | 18:01 |
| *** alexpilotti has quit IRC | 18:02 | |
| cinerama | yeah, i wanted to deliberate a bit over whether we could do things more properly without a big impact but i think on the whole better to do this for now and revisit it soon | 18:02 |
| *** alexpilotti has joined #openstack-ironic | 18:03 | |
| *** larsks has joined #openstack-ironic | 18:05 | |
| *** dtantsur is now known as dtantsur|afk | 18:07 | |
| *** d0ugal has quit IRC | 18:08 | |
| *** electrofelix has quit IRC | 18:09 | |
| *** rcernin has quit IRC | 18:09 | |
| *** jistr has joined #openstack-ironic | 18:15 | |
| *** jistr is now known as jistr|afk | 18:15 | |
| mat128 | thanks for your help with infra, JayF | 18:16 |
| JayF | np | 18:16 |
| mat128 | I posted on their forum for mirroring info, will follow up on this when I get a reply | 18:17 |
| *** openstackgerrit has quit IRC | 18:18 | |
| parlos | Good night | 18:18 |
| *** openstackgerrit has joined #openstack-ironic | 18:18 | |
| JayF | nn | 18:18 |
| *** ChrisAusten has joined #openstack-ironic | 18:22 | |
| *** parlos has quit IRC | 18:22 | |
| *** jtomasek has joined #openstack-ironic | 18:26 | |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Add pycrypto to requirements https://review.openstack.org/315213 | 18:26 |
| *** e0ne has quit IRC | 18:41 | |
| *** dmk0202 has joined #openstack-ironic | 18:47 | |
| *** konetzed has joined #openstack-ironic | 18:55 | |
| *** sambetts|afk has quit IRC | 18:59 | |
| jroll | JayF: what do you mean by infra liaison, is that a thing? | 19:02 |
| * jroll not sure it is | 19:02 | |
| JayF | yes | 19:02 |
| JayF | https://wiki.openstack.org/wiki/CrossProjectLiaisons#Infra | 19:02 |
| *** d0ugal has joined #openstack-ironic | 19:02 | |
| jroll | ah | 19:02 |
| *** alexpilotti has quit IRC | 19:02 | |
| jroll | is me acking those things typically a bottleneck? | 19:02 |
| *** alexpilotti has joined #openstack-ironic | 19:02 | |
| *** sambetts has joined #openstack-ironic | 19:03 | |
| JayF | I don't know, I asked before I looked up what it meant | 19:03 |
| JayF | I know liasons existed for all the other cross project stuff, so I thought we might want something there to, if not, that's ok too | 19:04 |
| jroll | heh | 19:04 |
| jroll | yeah I don't feel like I need to delegate that but will keep it in mind | 19:04 |
| *** fragatina has quit IRC | 19:08 | |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add node status checks to FPC sensors https://review.openstack.org/315725 | 19:12 |
| mat128 | JayF: Just got a reply from Tiny Core people: "Mirroring is not encouraged due to bandwidth issues." | 19:12 |
| mat128 | O_o | 19:12 |
| mat128 | http://forum.tinycorelinux.net/index.php/topic,19955.0.html | 19:13 |
| JayF | there's a better response now | 19:13 |
| mat128 | yeah | 19:13 |
| JayF | from the admin | 19:13 |
| mat128 | I'll try to see if rsync is public and open for their main site | 19:14 |
| JayF | well, they asked to pull down the initial sync from a mirror, which is reasonable | 19:15 |
| mat128 | yeah | 19:15 |
| mat128 | I'm more used to Debian or Ubuntu with push notifications and ftpsync | 19:16 |
| mat128 | but they seem to be like "well it's a simple rsync" | 19:16 |
| mat128 | true to the Tiny Core paradigm | 19:16 |
| JayF | well also I think complexity in mirroring is needed at larger distros | 19:17 |
| JayF | but they probably haven't hit that scaling point y et | 19:17 |
| mat128 | yup | 19:17 |
| JayF | I ran a freedos mirror for a long time; it was simple rsync for a while but eventually it got popular enough that I had to nuke it when they were chanigng how they synced it | 19:17 |
| mat128 | interesting | 19:21 |
| mat128 | TIL freedos had mirrors | 19:21 |
| mat128 | :) | 19:21 |
| JayF | FreeDOS powers many steam games (dosbox uses freedos so anything using dosbox uses freedos) | 19:24 |
| mat128 | Oh nice | 19:24 |
| * mat128 has certainly used dosbox in the past :) | 19:24 | |
| JayF | and the site I hosted the mirror on was the (now dead / unupdated for a decade) oldos.org | 19:25 |
| JayF | so it fit fairly well | 19:25 |
| *** r-daneel has joined #openstack-ironic | 19:28 | |
| *** athomas has quit IRC | 19:28 | |
| NobodyCam | vdrok: are you working on the back port of https://review.openstack.org/#/c/292552 .. I can if your busy? | 19:31 |
| *** Sukhdev has quit IRC | 19:32 | |
| mat128 | JayF: nice site, nostalgia just kicked in | 19:32 |
| *** Sukhdev has joined #openstack-ironic | 19:32 | |
| NobodyCam | JayF: now I want to play zork too | 19:33 |
| wajdi | oh wow....zork...blast from the past. | 19:34 |
| NobodyCam | lol | 19:35 |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Add some docs/comments to devstack/plugin.sh https://review.openstack.org/315740 | 19:37 |
| *** fragatina has joined #openstack-ironic | 19:46 | |
| *** mbound has joined #openstack-ironic | 19:49 | |
| *** jistr|afk has quit IRC | 19:52 | |
| *** sacharya_ has joined #openstack-ironic | 19:53 | |
| *** sacharya has quit IRC | 19:53 | |
| *** sacharya has joined #openstack-ironic | 19:53 | |
| *** sacharya_ has quit IRC | 19:54 | |
| *** ayoung has quit IRC | 20:20 | |
| *** wajdi has quit IRC | 20:20 | |
| *** keedya has quit IRC | 20:22 | |
| *** Sukhdev has quit IRC | 20:23 | |
| *** Sukhdev has joined #openstack-ironic | 20:24 | |
| *** rcernin has joined #openstack-ironic | 20:24 | |
| vdrok | NobodyCam: it's already on review :) | 20:27 |
| openstackgerrit | Jarrod Johnson proposed openstack/pyghmi: Add node status checks to FPC sensors https://review.openstack.org/315725 | 20:29 |
| NobodyCam | wow I totally missed it | 20:30 |
| *** rama_y has quit IRC | 20:31 | |
| *** hamzy has quit IRC | 20:36 | |
| JayF | mariojv: the allow vendor drivers to acquire shared locks, I can't remember why we wanted that originally | 20:39 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Fix all possible node subresources routing collisions https://review.openstack.org/315766 | 20:39 |
| JayF | mariojv: and I'm reviewing it now so thought it might be a good thing to know :) | 20:39 |
| mariojv | JayF: some folks methods don't require an exclusive lock | 20:40 |
| openstackgerrit | Sam Betts proposed openstack/ironic: Fix all possible node subresources routing collisions https://review.openstack.org/315766 | 20:40 |
| JayF | but they all take one now | 20:40 |
| JayF | got it | 20:40 |
| mariojv | sambetts was really interested in the patch iirc | 20:40 |
| *** raildo is now known as raildo-afk | 20:41 | |
| *** rama_y has joined #openstack-ironic | 20:44 | |
| *** Sukhdev has quit IRC | 20:46 | |
| jroll | JayF: yeah, the original thing was "don't make heartbeat lock" | 20:47 |
| rloo | JayF, cinerama: fwiw, I am uncomfortable adding pycrypto as a requirement in ironic, if it is only needed for ansible. | 20:47 |
| JayF | that would actaully be a really nice patch | 20:47 |
| JayF | rloo: I talked to them for a bit; apparently all other ansible deps are already enumerated in there | 20:47 |
| JayF | rloo: and so they changed the gate breakage bug to be "we handle dependencies badly" with a promise to fix | 20:47 |
| rloo | JayF: yeah, i saw that. I don't know what those other ones were added either. | 20:47 |
| rloo | JayF, cinerama: would you please, if it isn't commented in the file, add a comment indicating which ones were added only for ansible? | 20:48 |
| JayF | rloo: I'm just a reviewer, but I think that's a stellar idea | 20:48 |
| rloo | not everyone that uses ironic is using ansible. it isn't right to pull those other packages. | 20:48 |
| JayF | rloo: but that's for bifrost, which uses ansible | 20:49 |
| JayF | rloo: so it at least makes some sense at all | 20:49 |
| cinerama | this is for bifrost only | 20:50 |
| *** Goneri has quit IRC | 20:50 | |
| rloo | well, bifrost itself should make sure those packages are available then. | 20:51 |
| cinerama | we are, but due to the nature of how we install ansible this is how we currently do it | 20:52 |
| JayF | https://review.openstack.org/#/c/314739 easy ironic-spec to land | 20:53 |
| mrda | Morning Ironic | 20:54 |
| *** dprince has quit IRC | 20:54 | |
| jroll | rloo: have people asked for completed priorities to be documented? | 20:55 |
| jroll | or rather, what's the use in documenting such a thing there | 20:55 |
| jroll | we also have the RFEs and the spec location to document that | 20:55 |
| JayF | jroll: I liked it a lot, having a look-back to the last time and seeing how much we got done, helps set expectations better for *current* priorities | 20:55 |
| jroll | JayF: kinda late for that no? :) | 20:56 |
| rloo | jroll: that is just me. if you have priorities, it seems good to know how well we did. | 20:56 |
| rloo | one could look at specs etc, but that is digging around for stuff. | 20:56 |
| rloo | if no one else wants it, just -1. | 20:56 |
| jroll | rloo: yeah, not opposed to it, just curious why | 20:56 |
| rloo | maybe partly for accountability. or visibility to folks so they can see what we wanted to do and what actually got done. | 20:56 |
| JayF | I can't imagine why anyone wouldn't want something documented given it's time bound and will never be outta date | 20:57 |
| mrda | JayF: people forget, that's one reason why we document :) | 20:58 |
| jroll | JayF: more like it's already documented in two places :P | 20:58 |
| mat128 | To go along with JayF, that's the only thing we can document that will never rot | 20:58 |
| mat128 | but we should be able to derive this from the things that we committed along the cycle | 20:59 |
| mat128 | release notes maybe? | 20:59 |
| JayF | jroll: yeah, I think that's a good argument if the doc would rot, but given it won't, I don't see the harm | 20:59 |
| jroll | it'd be in release notes, launchpad (via RFEs), specs/implemented/mitaka/*, hm what elese | 21:00 |
| jroll | but yeah I landed it | 21:00 |
| *** ChrisAusten has quit IRC | 21:00 | |
| rloo | for those following that conversation with jroll, this is what he meant: https://review.openstack.org/#/c/314739/ | 21:01 |
| rloo | oh, I didn't notice that JayF had brought it up :) | 21:02 |
| jroll | heh | 21:02 |
| rloo | so fwiw, I'd really like to get the node adoption stuff in. i just reviewed but had one big issue with it.https://review.openstack.org/#/c/275766 | 21:03 |
| JayF | I always link something in here when I +2 or +1 it and all it needs is another core and it's simple | 21:03 |
| openstackgerrit | Merged openstack/ironic-specs: Priorities that were done in Mitaka https://review.openstack.org/314739 | 21:03 |
| rloo | jroll: ^^ I might add it to monday's agenda to discuss, just to resolve if there is contention. i don't want to do reviews that span over many months. let's just knock them out. | 21:03 |
| *** yolanda has quit IRC | 21:04 | |
| *** yolanda has joined #openstack-ironic | 21:04 | |
| *** baoli_ has quit IRC | 21:10 | |
| *** cdearborn has quit IRC | 21:10 | |
| JayF | rloo: re https://review.openstack.org/#/c/275766, Julia is out for at least another week; if we want to get it landed someone will have to update the patch | 21:11 |
| *** xek has quit IRC | 21:11 | |
| devananda | rloo: ++ on not adding pycrypto for ansible deploy driver | 21:12 |
| rloo | JayF: yeah, i have no problems with someone else updating it. i am fine doing a follow on for the doc stuff i pointed out. it is just the one issue that we need to decide on. i mean, that i want people to agree with me. ha ha. | 21:12 |
| *** xek has joined #openstack-ironic | 21:12 | |
| devananda | rloo: we already have a separate file to track driver dependencies | 21:12 |
| jroll | were we talking about ansible driver or bifrost? I thought it was bifrost | 21:13 |
| JayF | devananda: the patch is to bifrost | 21:13 |
| rloo | devananda: w/o that, i think bifrost's gate is broken | 21:13 |
| rloo | JayF: OH. I thought the patch was in ironic. | 21:13 |
| devananda | JayF: oh! | 21:13 |
| devananda | hmm. why is the gate broken? shouldn't ansible be getting installed -- along with whatever deps it has? | 21:13 |
| cinerama | hee hee | 21:13 |
| devananda | cinerama: "the nature of how we install ansible" -- could you elaborate? | 21:13 |
| rloo | oh oh. devananda has now gotten suckered into that rat hole. | 21:14 |
| devananda | haha | 21:14 |
| cinerama | devananda, this is in the bug, but we run ansible from a checkout by default | 21:14 |
| rloo | JayF: wrt TheJulia's patch, I'm going to add it as an item in Monday's meeting. (unless it gets resolved before then.) | 21:14 |
| JayF | heh, and now cinerama gets to have this conversatino for the 3rd time :P | 21:14 |
| cinerama | devananda, so pip doesn't actually touch it. in our bootstrap script we install deps from the requirements file using pip prior to running ansible | 21:14 |
| cinerama | JayF: i'm well rehearsed now :) | 21:15 |
| devananda | cinerama: well, installing things that aren't coming from openstack mirrors is a frequent cause of gate breakages | 21:15 |
| devananda | and increaed challenges in debugging said gate issues | 21:16 |
| devananda | so, what is the reason to not install ansible's latest release? | 21:16 |
| cinerama | devananda: so we might change how we install in a while but at the moment i want to a) go for the principle of least surprise from the user perspective and b) fix the gate | 21:16 |
| cinerama | this is historically how we've done it, i suspect in part because at some points we've run off the devel branch | 21:18 |
| *** Sukhdev has joined #openstack-ironic | 21:18 | |
| cinerama | so we don't want necessarily want to tread on the user's existing ansible install if they have one | 21:18 |
| jroll | so this is how it's installed, in a venv https://github.com/openstack/bifrost/blob/master/scripts/env-setup.sh | 21:19 |
| devananda | cinerama: are we talking about the gate and just the gate? or how bifrost is installed in general? | 21:19 |
| jroll | (I think in a venv) | 21:19 |
| jroll | I actually can't tell | 21:20 |
| cinerama | jroll, you can pick whether you're running in a venv or if you're using a checkout + ansible's "hacking" env setup | 21:20 |
| jroll | ah | 21:20 |
| openstackgerrit | Merged openstack/pyghmi: Add node status checks to FPC sensors https://review.openstack.org/315725 | 21:20 |
| jroll | default is not venv? | 21:20 |
| jroll | oh, ansible hacking thing is a venv-like thing | 21:21 |
| cinerama | yep | 21:21 |
| jroll | so they have this: https://github.com/ansible/ansible/blob/devel/setup.py#L25 | 21:22 |
| cinerama | yeah they don't use a requirements file | 21:22 |
| jroll | these are almost certainly installed with "pip install /opt/stack/ansible" | 21:22 |
| jroll | but I guess not in the hacking thing? | 21:22 |
| cinerama | that's right. we never actually pip install that source tree if we're not installing to a venv | 21:22 |
| *** cdearborn has joined #openstack-ironic | 21:23 | |
| jroll | hrm | 21:24 |
| jroll | is there any reason *not* to use virtualenv? | 21:24 |
| cinerama | i think it could be good but i don't want to change the way we behave drastically right now | 21:26 |
| jroll | sure | 21:27 |
| jroll | so I'm totally +1 on unbreaking the gate by adding pycrypto | 21:27 |
| jroll | and happy to push a button on a patch | 21:27 |
| jroll | but I do think it should be reworked | 21:27 |
| jroll | (is there a patch?) | 21:27 |
| JayF | jroll: yeah, I made them file a bug about it before I pulled me -1 from the patch | 21:27 |
| cinerama | i agree and that's why i updated the bug | 21:27 |
| JayF | jroll: https://review.openstack.org/#/c/315213/ | 21:27 |
| cinerama | so the patch is now partial bug | 21:27 |
| NobodyCam | jroll: there is a bug.. like JayF just said | 21:27 |
| jroll | perf | 21:27 |
| NobodyCam | :) | 21:28 |
| JayF | I had this all handled and then three other people came along and handled it again, lol | 21:28 |
| devananda | :) | 21:28 |
| jroll | cinerama: NobodyCam: I +2'd if you want to land it | 21:28 |
| jroll | JayF: lol | 21:28 |
| jroll | just making sure I understand | 21:28 |
| jroll | I did go read code at least :) | 21:28 |
| NobodyCam | ack doing now | 21:29 |
| jroll | cinerama: so I think it's just a few lines of python code to pull out the setup_requires stuff | 21:29 |
| jroll | which I think would be a fine fix for the hacking/env-setup model | 21:29 |
| cinerama | oh yeah there are lots of things we could do to fix | 21:30 |
| cinerama | i just want to chat with julia about what the best thing to do with install of ansible in general is in context of the overall vision | 21:30 |
| NobodyCam | +2/+a with a comment | 21:30 |
| jroll | totally | 21:31 |
| jroll | cinerama: sorry for the multi-day run around here :( | 21:31 |
| cinerama | oh no worries | 21:31 |
| *** ChrisAusten has joined #openstack-ironic | 21:33 | |
| JayF | jroll: https://review.openstack.org/#/c/314768 is kinda a policy thing, looks like you should review, but it's pretty simple if you agree and wanna land | 21:33 |
| *** ametts has quit IRC | 21:33 | |
| *** dtantsur|afk has quit IRC | 21:33 | |
| jroll | hm, that's kind of chicken and egg no? | 21:34 |
| jroll | you can't prove your system is reliable until the code is in | 21:34 |
| jroll | krtaylor: you here or should I comment on the patch there? | 21:34 |
| JayF | jroll: well I thought about that; but you can def setup third party ci that would install a downstream driver and run it | 21:35 |
| JayF | jroll: and I presume if someone couldn't do that for $reasons, they'd ask on ML or here and we'd be our usual, reasonable selves | 21:35 |
| *** dtantsur has joined #openstack-ironic | 21:35 | |
| jroll | JayF: some might take the documentation to heart and nack a spec for a new driver because there isn't CI | 21:36 |
| JayF | maybe the fairer way would be "must have voting 3rd party CI before the driver is *released*" or similar? | 21:36 |
| jroll | yeah, needs more wording somehow | 21:37 |
| jlvillal | So in devstack. There is 'IRONIC_BAREMETAL_BASIC_OPS' Trying to figure out purpose of it. Like when can it be set to False if we are using Ironic. | 21:41 |
| * jlvillal feels like this is a question he should already know the answer too | 21:41 | |
| jroll | jlvillal: did you check devstack for it? | 21:42 |
| jlvillal | jroll: I'm walking through the lib/ironic code now. | 21:42 |
| jlvillal | Reading it. | 21:42 |
| jlvillal | ironic/devstack/lib/ironic code. | 21:42 |
| jroll | jlvillal: oh, it isn't in devstack, just our plugin | 21:42 |
| jroll | seems we do extra things when that's set | 21:42 |
| jroll | I think the idea is that you're going to run tempest against this thing? idk | 21:43 |
| jlvillal | Right. I'm just wondering when would we not set it. If that is valid. | 21:43 |
| jroll | yeah idk | 21:43 |
| jlvillal | Wondering if it is cruft or a valid case to be False | 21:43 |
| jroll | it enrolls nodes and stuff | 21:43 |
| jroll | so I guess if you deployed prod on devstack maybe it makes sense a bit? | 21:43 |
| JayF | our gate fix will merge any second now, fwiw | 21:44 |
| jroll | handle your own enrollment, ramdisk, tftpd, etc | 21:44 |
| jroll | idk, can always send an email | 21:44 |
| jlvillal | jroll: Thanks. Just trying to dig into this devstack stuff and comment it while I'm going along. To then understand better the grenade stuff where it is failing. | 21:44 |
| openstackgerrit | Merged openstack/bifrost: Add pycrypto to requirements https://review.openstack.org/315213 | 21:44 |
| jlvillal | jroll: And thanks again for figuring out the neutron issue with 'resource phase create'! :) | 21:45 |
| jroll | jlvillal: well, it isn't failing in devstack, right? | 21:45 |
| jroll | np | 21:45 |
| jroll | argh so many smoke test timeouts :| | 21:45 |
| jlvillal | JayF: It disappeared from my Zuul board | 21:46 |
| jlvillal | JayF: And merged :) https://review.openstack.org/#/c/312751/ | 21:46 |
| JayF | yep just landed | 21:46 |
| jlvillal | Hopefully that fixes the gate | 21:46 |
| JayF | when I looked just then it was waiting on one job almost done right above it | 21:46 |
| JayF | I'll reverify some stuff | 21:46 |
| *** dprince has joined #openstack-ironic | 21:49 | |
| dansmith | I hereby apologize to ironic people | 21:50 |
| * NobodyCam raises a eyebrow | 21:51 | |
| JayF | shit happens, we just need to finish stuffing the agent in a smaller container so we can run better ci in nova :) | 21:51 |
| JayF | NobodyCam: I saw dansmith with bolt cutters near the gate :P | 21:51 |
| NobodyCam | oh :) | 21:51 |
| dansmith | literally like two lines that ended up rebased into the wrong patch created something that didn't work out of order (for ironic) | 21:51 |
| dansmith | my bad :) | 21:51 |
| *** ChrisAusten has quit IRC | 21:53 | |
| rloo | dansmith: apology accepted! Just be more careful in the future, ok? :) | 21:54 |
| jlvillal | Thanks dansmith for fixing it | 21:54 |
| dansmith | jlvillal: I didn't fix it, the patch with the other half of the ordering merged :) | 21:54 |
| dansmith | rloo: yeah, something something fix your gate job something :) | 21:54 |
| jlvillal | Potato potato (doesn't quite work in a text only medium...) | 21:55 |
| jroll | potato potahto | 21:55 |
| jlvillal | :) | 21:55 |
| JayF | if someone said potato to me 6 times in a row with no explanation, I'd want to call the whole thing off too | 21:55 |
| rloo | pota-toe potahto | 21:55 |
| cinerama | or "to MAH toe sauce" vs "to MAY toe sauce" :) | 22:00 |
| *** dmk0202 has quit IRC | 22:00 | |
| *** alexpilotti has quit IRC | 22:01 | |
| *** alexpilo_ has joined #openstack-ironic | 22:01 | |
| * jroll out, see y'all tomorrow | 22:01 | |
| NobodyCam | night jroll ... | 22:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/bifrost: Updated from global requirements https://review.openstack.org/313712 | 22:02 |
| * JayF out, off tomorrow excepting a couple of hours | 22:05 | |
| NobodyCam | night JayF | 22:06 |
| *** Sukhdev has quit IRC | 22:07 | |
| *** cdearborn has quit IRC | 22:13 | |
| *** ayoung has joined #openstack-ironic | 22:25 | |
| *** earlephilhower has joined #openstack-ironic | 22:25 | |
| *** absubram has quit IRC | 22:28 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Only install DIB if going to use DIB https://review.openstack.org/315795 | 22:29 |
| *** rloo has quit IRC | 22:31 | |
| *** rama_y has quit IRC | 22:46 | |
| *** dprince has quit IRC | 22:57 | |
| *** dprince has joined #openstack-ironic | 22:57 | |
| *** thiagop has quit IRC | 22:59 | |
| *** ayoung has quit IRC | 23:03 | |
| *** Sukhdev has joined #openstack-ironic | 23:09 | |
| *** intr1nsic has quit IRC | 23:13 | |
| *** intr1nsic has joined #openstack-ironic | 23:15 | |
| *** wajdi has joined #openstack-ironic | 23:18 | |
| *** rloo has joined #openstack-ironic | 23:19 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Only install DIB if going to use DIB https://review.openstack.org/315795 | 23:20 |
| *** yuanying_ has joined #openstack-ironic | 23:22 | |
| *** yuanying has quit IRC | 23:23 | |
| *** garthb__ has quit IRC | 23:25 | |
| *** r-daneel has quit IRC | 23:27 | |
| *** rloo has quit IRC | 23:34 | |
| *** chlong has quit IRC | 23:35 | |
| *** Sukhdev has quit IRC | 23:48 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!